Publications by type
For a complete list of publications in chronological order see this page.
Selected international journals
1.
Łukasz Radliński
The Impact of Data Quality on Software Testing Effort Prediction
In: Electronics, vol. 12, no. 7, pp. 1656, 2023, ISSN: 2079-9292.
@article{Radlinski_2023_electronics,
title = {The Impact of Data Quality on Software Testing Effort Prediction},
author = {Łukasz Radliński},
url = {https://www.mdpi.com/2079-9292/12/7/1656},
doi = {10.3390/electronics12071656},
issn = {2079-9292},
year = {2023},
date = {2023-03-01},
journal = {Electronics},
volume = {12},
number = {7},
pages = {1656},
abstract = {Background: This paper investigates the impact of data quality on the performance of models predicting effort on software testing. Data quality was reflected by training data filtering strategies (data variants) covering combinations of Data Quality Rating, UFP Rating, and a threshold of valid cases. Methods: The experiment used the ISBSG dataset and 16 machine learning models. A process of three-fold cross-validation repeated 20 times was used to train and evaluate each model with each data variant. Model performance was assessed using absolute errors of prediction. A ‘win–tie–loss' procedure, based on the Wilcoxon signed-rank test, was applied to identify the best models and data variants. Results: Most models, especially the most accurate, performed the best on a complete dataset, even though it contained cases with low data ratings. The detailed results include the rankings of the following: (1) models for particular data variants, (2) data variants for particular models, and (3) the best-performing combinations of models and data variants. Conclusions: Arbitrary and restrictive data selection to only projects with Data Quality Rating and UFP Rating of ‘A' or ‘B', commonly used in the literature, does not seem justified. It is recommended not to exclude cases with low data ratings to achieve better accuracy of most predictive models for testing effort prediction.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Background: This paper investigates the impact of data quality on the performance of models predicting effort on software testing. Data quality was reflected by training data filtering strategies (data variants) covering combinations of Data Quality Rating, UFP Rating, and a threshold of valid cases. Methods: The experiment used the ISBSG dataset and 16 machine learning models. A process of three-fold cross-validation repeated 20 times was used to train and evaluate each model with each data variant. Model performance was assessed using absolute errors of prediction. A ‘win–tie–loss' procedure, based on the Wilcoxon signed-rank test, was applied to identify the best models and data variants. Results: Most models, especially the most accurate, performed the best on a complete dataset, even though it contained cases with low data ratings. The detailed results include the rankings of the following: (1) models for particular data variants, (2) data variants for particular models, and (3) the best-performing combinations of models and data variants. Conclusions: Arbitrary and restrictive data selection to only projects with Data Quality Rating and UFP Rating of ‘A' or ‘B', commonly used in the literature, does not seem justified. It is recommended not to exclude cases with low data ratings to achieve better accuracy of most predictive models for testing effort prediction.
2.
Anthony Costa Constantinou; Norman Fenton; William Marsh; Lukasz Radlinski
In: Artificial Intelligence in Medicine, vol. 67, pp. 75–93, 2016, ISSN: 09333657.
@article{Constantinou2016,
title = {From complex questionnaire and interviewing data to intelligent Bayesian network models for medical decision support},
author = {Anthony Costa Constantinou and Norman Fenton and William Marsh and Lukasz Radlinski},
url = {http://linkinghub.elsevier.com/retrieve/pii/S093336571600004X},
doi = {10.1016/j.artmed.2016.01.002},
issn = {09333657},
year = {2016},
date = {2016-02-01},
journal = {Artificial Intelligence in Medicine},
volume = {67},
pages = {75–93},
abstract = {Objectives (1) To develop a rigorous and repeatable method for building effective Bayesian network (BN) models for medical decision support from complex, unstructured and incomplete patient questionnaires and interviews that inevitably contain examples of repetitive, redundant and contradictory responses; (2) To exploit expert knowledge in the BN development since further data acquisition is usually not possible; (3) To ensure the BN model can be used for interventional analysis; (4) To demonstrate why using data alone to learn the model structure and parameters is often unsatisfactory even when extensive data is available. Method The method is based on applying a range of recent BN developments targeted at helping experts build BNs given limited data. While most of the components of the method are based on established work, its novelty is that it provides a rigorous consolidated and generalised framework that addresses the whole life-cycle of BN model development. The method is based on two original and recent validated BN models in forensic psychiatry, known as DSVM-MSS and DSVM-P. Results When employed with the same datasets, the DSVM-MSS demonstrated competitive to superior predictive performance (AUC scores 0.708 and 0.797) against the state-of-the-art (AUC scores ranging from 0.527 to 0.705), and the DSVM-P demonstrated superior predictive performance (cross-validated AUC score of 0.78) against the state-of-the-art (AUC scores ranging from 0.665 to 0.717). More importantly, the resulting models go beyond improving predictive accuracy and into usefulness for risk management purposes through intervention, and enhanced decision support in terms of answering complex clinical questions that are based on unobserved evidence. Conclusions This development process is applicable to any application domain which involves large-scale decision analysis based on such complex information, rather than based on data with hard facts, and in conjunction with the incorporation of expert knowledge for decision support via intervention. The novelty extends to challenging the decision scientists to reason about building models based on what information is really required for inference, rather than based on what data is available and hence, forces decision scientists to use available data in a much smarter way.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Objectives (1) To develop a rigorous and repeatable method for building effective Bayesian network (BN) models for medical decision support from complex, unstructured and incomplete patient questionnaires and interviews that inevitably contain examples of repetitive, redundant and contradictory responses; (2) To exploit expert knowledge in the BN development since further data acquisition is usually not possible; (3) To ensure the BN model can be used for interventional analysis; (4) To demonstrate why using data alone to learn the model structure and parameters is often unsatisfactory even when extensive data is available. Method The method is based on applying a range of recent BN developments targeted at helping experts build BNs given limited data. While most of the components of the method are based on established work, its novelty is that it provides a rigorous consolidated and generalised framework that addresses the whole life-cycle of BN model development. The method is based on two original and recent validated BN models in forensic psychiatry, known as DSVM-MSS and DSVM-P. Results When employed with the same datasets, the DSVM-MSS demonstrated competitive to superior predictive performance (AUC scores 0.708 and 0.797) against the state-of-the-art (AUC scores ranging from 0.527 to 0.705), and the DSVM-P demonstrated superior predictive performance (cross-validated AUC score of 0.78) against the state-of-the-art (AUC scores ranging from 0.665 to 0.717). More importantly, the resulting models go beyond improving predictive accuracy and into usefulness for risk management purposes through intervention, and enhanced decision support in terms of answering complex clinical questions that are based on unobserved evidence. Conclusions This development process is applicable to any application domain which involves large-scale decision analysis based on such complex information, rather than based on data with hard facts, and in conjunction with the incorporation of expert knowledge for decision support via intervention. The novelty extends to challenging the decision scientists to reason about building models based on what information is really required for inference, rather than based on what data is available and hence, forces decision scientists to use available data in a much smarter way.
3.
Thomas Schulz; Łukasz Radliński; Thomas Gorges; Wolfgang Rosenstiel
Predicting the Flow of Defect Correction Effort using a Bayesian Network Model
In: Empirical Software Engineering, vol. 18, no. 3, pp. 435–477, 2013, ISSN: 1382-3256.
@article{Schulz_emse2013,
title = {Predicting the Flow of Defect Correction Effort using a Bayesian Network Model},
author = {Thomas Schulz and Łukasz Radliński and Thomas Gorges and Wolfgang Rosenstiel},
url = {http://www.springerlink.com/index/10.1007/s10664-011-9175-7},
doi = {10.1007/s10664-011-9175-7},
issn = {1382-3256},
year = {2013},
date = {2013-09-01},
journal = {Empirical Software Engineering},
volume = {18},
number = {3},
pages = {435–477},
abstract = {The number of defects alone does not provide software companies with enough information on the effort required to fix them. Defects have different impacts on the overall defect correction effort – defects introduced in one phase may be found and corrected in the same or later phase. The later they are found, the more effort is required to correct them. The main aim of this paper is to build and validate a model (Bayesian Network) for predicting the defect correction effort at various phases of the software development process. The procedure of building the model covers the following steps: problem analysis, data analysis, model definition and enhancement, simulation runs, and model validation. Developed Defect Cost Flow Model (DCFM), which is an implementation of the V-model of a software project lifecycle, correctly incorporates known qualitative and quantitative relationships. Application of DCFM in a real industrial process revealed its high potential in finding the appropriate amount of review effort for specific development phases to minimize the overall costs. The model may be used in the industry for decision support. It can be extended and calibrated to meet the needs of specific development environment.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The number of defects alone does not provide software companies with enough information on the effort required to fix them. Defects have different impacts on the overall defect correction effort – defects introduced in one phase may be found and corrected in the same or later phase. The later they are found, the more effort is required to correct them. The main aim of this paper is to build and validate a model (Bayesian Network) for predicting the defect correction effort at various phases of the software development process. The procedure of building the model covers the following steps: problem analysis, data analysis, model definition and enhancement, simulation runs, and model validation. Developed Defect Cost Flow Model (DCFM), which is an implementation of the V-model of a software project lifecycle, correctly incorporates known qualitative and quantitative relationships. Application of DCFM in a real industrial process revealed its high potential in finding the appropriate amount of review effort for specific development phases to minimize the overall costs. The model may be used in the industry for decision support. It can be extended and calibrated to meet the needs of specific development environment.
4.
Lukasz Radlinski
A Survey of Bayesian Net Models for Software Development Effort Prediction
In: International Journal of Software Engineering and Computing, vol. 2, no. 2, pp. 95–109, 2010.
@article{Radlinski_ijsec_bns2010,
title = {A Survey of Bayesian Net Models for Software Development Effort Prediction},
author = {Lukasz Radlinski},
year = {2010},
date = {2010-01-01},
journal = {International Journal of Software Engineering and Computing},
volume = {2},
number = {2},
pages = {95–109},
abstract = {This paper discusses recent Bayesian nets built for software development effort prediction. Its aim is to bring closer these models as they may be competitive for other modeling techniques, especially for data-driven machine learning and statistical techniques. Each model has been briefly described and then analyzed in detail in terms of its main purpose, type of structure, data/knowledge base for building a model. Some models have been empirically validated for predictive accuracy and we discuss the results of this validation. The paper also discusses main problems related to building such models by domain experts.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This paper discusses recent Bayesian nets built for software development effort prediction. Its aim is to bring closer these models as they may be competitive for other modeling techniques, especially for data-driven machine learning and statistical techniques. Each model has been briefly described and then analyzed in detail in terms of its main purpose, type of structure, data/knowledge base for building a model. Some models have been empirically validated for predictive accuracy and we discuss the results of this validation. The paper also discusses main problems related to building such models by domain experts.
5.
Lukasz Radlinski; Wladyslaw Hoffmann
On Predicting Software Development Effort using Machine Learning Techniques and Local Data
In: International Journal of Software Engineering and Computing, vol. 2, no. 2, pp. 123–136, 2010.
@article{Radlinski_ijsec_mleffort2010,
title = {On Predicting Software Development Effort using Machine Learning Techniques and Local Data},
author = {Lukasz Radlinski and Wladyslaw Hoffmann},
year = {2010},
date = {2010-01-01},
journal = {International Journal of Software Engineering and Computing},
volume = {2},
number = {2},
pages = {123–136},
abstract = {This paper analyses the accuracy of predictions for software development effort using various machine learning techniques. The main aim is to investigate the stability of these predictions by analyzing if particular techniques achieve a similar level of accuracy for different datasets. Two key assumptions are that (1) predictions are performed using local empirical data and (2) very little expert input is required. The study involves using 23 machine learning techniques with four publicly available datasets: COCOMO, Desharnais, Maxwell and QQDefects. The results show that the accuracy of predictions for each technique varies depending on the dataset used. With feature selection most techniques provide higher predictive accuracy and this accuracy is more stable across different datasets. The highest positive impact of feature selection on the accuracy has been observed for the K* technique, which has generated the most accurate predictions across all datasets.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This paper analyses the accuracy of predictions for software development effort using various machine learning techniques. The main aim is to investigate the stability of these predictions by analyzing if particular techniques achieve a similar level of accuracy for different datasets. Two key assumptions are that (1) predictions are performed using local empirical data and (2) very little expert input is required. The study involves using 23 machine learning techniques with four publicly available datasets: COCOMO, Desharnais, Maxwell and QQDefects. The results show that the accuracy of predictions for each technique varies depending on the dataset used. With feature selection most techniques provide higher predictive accuracy and this accuracy is more stable across different datasets. The highest positive impact of feature selection on the accuracy has been observed for the K* technique, which has generated the most accurate predictions across all datasets.
6.
Norman Fenton; Martin Neil; William Marsh; Peter Hearty; Łukasz Radliński; Paul Krause
On the effectiveness of early life cycle defect prediction with Bayesian Nets
In: Empirical Software Engineering, vol. 13, no. 5, pp. 499 –537, 2008.
@article{Fenton_emse2008,
title = {On the effectiveness of early life cycle defect prediction with Bayesian Nets},
author = {Norman Fenton and Martin Neil and William Marsh and Peter Hearty and Łukasz Radliński and Paul Krause},
url = {http://www.springerlink.com/content/kg06211161305k1t},
doi = {10.1007/s10664-008-9072-x},
year = {2008},
date = {2008-01-01},
journal = {Empirical Software Engineering},
volume = {13},
number = {5},
pages = {499 –537},
abstract = {Standard practice in building models in software engineering normally involves three steps: collecting domain knowledge (previous results, expert knowledge); building a skeleton of the model based on step 1 including as yet unknown parameters; estimating the model parameters using historical data. Our experience shows that it is extremely difficult to obtain reliable data of the required granularity, or of the required volume with which we could later generalize our conclusions. Therefore, in searching for a method for building a model we cannot consider methods requiring large volumes of data. This paper discusses an experiment to develop a causal model (Bayesian net) for predicting the number of residual defects that are likely to be found during independent testing or operational usage. The approach supports (1) and (2), does not require (3), yet still makes accurate defect predictions (an R 2 of 0.93 between predicted and actual defects). Since our method does not require detailed domain knowledge it can be applied very early in the process life cycle. The model incorporates a set of quantitative and qualitative factors describing a project and its development process, which are inputs to the model. The model variables, as well as the relationships between them, were identified as part of a major collaborative project. A dataset, elicited from 31 completed software projects in the consumer electronics industry, was gathered using a questionnaire distributed to managers of recent projects. We used this dataset to validate the model by analyzing several popular evaluation measures (R 2, measures based on the relative error and Pred). The validation results also confirm the need for using the qualitative factors in the model. The dataset may be of interest to other researchers evaluating models with similar aims. Based on some typical scenarios we demonstrate how the model can be used for better decision support in operational environments. We also performed sensitivity analysis in which we identified the most influential variables on the number of residual defects. This showed that the project size, scale of distributed communication and the project complexity cause the most of variation in number of defects in our model. We make both the dataset and causal model available for research use.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Standard practice in building models in software engineering normally involves three steps: collecting domain knowledge (previous results, expert knowledge); building a skeleton of the model based on step 1 including as yet unknown parameters; estimating the model parameters using historical data. Our experience shows that it is extremely difficult to obtain reliable data of the required granularity, or of the required volume with which we could later generalize our conclusions. Therefore, in searching for a method for building a model we cannot consider methods requiring large volumes of data. This paper discusses an experiment to develop a causal model (Bayesian net) for predicting the number of residual defects that are likely to be found during independent testing or operational usage. The approach supports (1) and (2), does not require (3), yet still makes accurate defect predictions (an R 2 of 0.93 between predicted and actual defects). Since our method does not require detailed domain knowledge it can be applied very early in the process life cycle. The model incorporates a set of quantitative and qualitative factors describing a project and its development process, which are inputs to the model. The model variables, as well as the relationships between them, were identified as part of a major collaborative project. A dataset, elicited from 31 completed software projects in the consumer electronics industry, was gathered using a questionnaire distributed to managers of recent projects. We used this dataset to validate the model by analyzing several popular evaluation measures (R 2, measures based on the relative error and Pred). The validation results also confirm the need for using the qualitative factors in the model. The dataset may be of interest to other researchers evaluating models with similar aims. Based on some typical scenarios we demonstrate how the model can be used for better decision support in operational environments. We also performed sensitivity analysis in which we identified the most influential variables on the number of residual defects. This showed that the project size, scale of distributed communication and the project complexity cause the most of variation in number of defects in our model. We make both the dataset and causal model available for research use.
Other journals
1.
Łukasz Radliński
Analysis of factors of software development effort and productivity
In: Procedia Computer Science, vol. 192, pp. 4790–4799, 2021, ISSN: 18770509.
@article{Radlinski_kes2021,
title = {Analysis of factors of software development effort and productivity},
author = {Łukasz Radliński},
url = {https://www.sciencedirect.com/science/article/pii/S1877050921019967},
doi = {10.1016/j.procs.2021.09.257},
issn = {18770509},
year = {2021},
date = {2021-01-01},
journal = {Procedia Computer Science},
volume = {192},
pages = {4790–4799},
abstract = {The goal of this paper was to identify factors of development effort and productivity and investigate the nature of these relationships using the current release of the ISBSG dataset. In particular, statistical measures of correlations and associations, single-predictor linear regression models, and, most importantly, a moderation analysis were used. Performed analysis demonstrated which attributes are in strong relationships with effort and productivity, investigated the explainability of single-predictor models, and discussed if and how particular attributes moderate the strongest relationships reflected in these single-predictor models.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The goal of this paper was to identify factors of development effort and productivity and investigate the nature of these relationships using the current release of the ISBSG dataset. In particular, statistical measures of correlations and associations, single-predictor linear regression models, and, most importantly, a moderation analysis were used. Performed analysis demonstrated which attributes are in strong relationships with effort and productivity, investigated the explainability of single-predictor models, and discussed if and how particular attributes moderate the strongest relationships reflected in these single-predictor models.
2.
Lukasz Radlinski
Stability of user satisfaction prediction in software projects
In: Procedia Computer Science, vol. 176, pp. 2394–2403, 2020, ISSN: 18770509.
@article{Radlinski2020_kes,
title = {Stability of user satisfaction prediction in software projects},
author = {Lukasz Radlinski},
url = {https://www.sciencedirect.com/science/article/pii/S1877050920322195},
doi = {10.1016/j.procs.2020.09.308},
issn = {18770509},
year = {2020},
date = {2020-01-01},
journal = {Procedia Computer Science},
volume = {176},
pages = {2394–2403},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3.
Łukasz Radliński
Predicting Aggregated User Satisfaction in Software Projects
In: Foundations of Computing and Decision Sciences, vol. 43, no. 4, pp. 335–357, 2018, ISSN: 2300-3405.
@article{Radlinski2018,
title = {Predicting Aggregated User Satisfaction in Software Projects},
author = {Łukasz Radliński},
url = {http://content.sciendo.com/view/journals/fcds/43/4/article-p335.xml},
doi = {10.1515/fcds-2018-0017},
issn = {2300-3405},
year = {2018},
date = {2018-12-01},
journal = {Foundations of Computing and Decision Sciences},
volume = {43},
number = {4},
pages = {335–357},
abstract = {User satisfaction is an important feature of software quality. However, it was rarely studied in software engineering literature. By enhancing earlier research this paper focuses on predicting user satisfaction with machine learning techniques using software development data from an extended ISBSG dataset. This study involved building, evaluating and comparing a total of 15,600 prediction schemes. Each scheme consists of a different combination of its components: manual feature preselection, handling missing values, outlier elimination, value normalization, automated feature selection, and a classifier. The research procedure involved a 10-fold cross-validation and separate testing, both repeated 10 times, to train and to evaluate each prediction scheme. Achieved level of accuracy for best performing schemes expressed by Matthews correlation coefficient was about 0.5 in the cross-validation and about 0.5–0.6 in the testing stage. The study identified the most accurate settings for components of prediction schemes.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
User satisfaction is an important feature of software quality. However, it was rarely studied in software engineering literature. By enhancing earlier research this paper focuses on predicting user satisfaction with machine learning techniques using software development data from an extended ISBSG dataset. This study involved building, evaluating and comparing a total of 15,600 prediction schemes. Each scheme consists of a different combination of its components: manual feature preselection, handling missing values, outlier elimination, value normalization, automated feature selection, and a classifier. The research procedure involved a 10-fold cross-validation and separate testing, both repeated 10 times, to train and to evaluate each prediction scheme. Achieved level of accuracy for best performing schemes expressed by Matthews correlation coefficient was about 0.5 in the cross-validation and about 0.5–0.6 in the testing stage. The study identified the most accurate settings for components of prediction schemes.
4.
Łukasz Radliński
In: Journal of Theoretical and Applied Computer Science, vol. 9, no. 2, pp. 32–50, 2015.
@article{Radlinski2015,
title = {Preliminary evaluation of schemes for predicting user satisfaction with the ability of system to meet stated objectives},
author = {Łukasz Radliński},
url = {http://www.jtacs.org/archive/2015/2/3},
year = {2015},
date = {2015-01-01},
journal = {Journal of Theoretical and Applied Computer Science},
volume = {9},
number = {2},
pages = {32–50},
abstract = {In software engineering literature two most commonly investigated targets for prediction are development effort and software quality. This study follows the methodological advances of these studies but focuses on predicting user satisfaction in software project. Specific outcome variable investigated in prediction is user satisfaction with the ability of system to meet stated objectives (MSO). A total number of 288 prediction schemes have been evaluated in the ability to predict MSO. These schemes have been built as different combinations of their components, i.e. feature pre-selection, elimination of missing values, automated feature selection, and a classifier. Two best performing schemes achieved the accuracy measured as Matthews correlation coefficient of 0.71 in test subset. These schemes involved W-LMT and W-SimpleLogistic classifiers. Significant differences have been observed between different classifiers and selected other components, depending on the dataset (validation or test). Discussed results may serve as guidelines to design a scheme to predict user satisfaction.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In software engineering literature two most commonly investigated targets for prediction are development effort and software quality. This study follows the methodological advances of these studies but focuses on predicting user satisfaction in software project. Specific outcome variable investigated in prediction is user satisfaction with the ability of system to meet stated objectives (MSO). A total number of 288 prediction schemes have been evaluated in the ability to predict MSO. These schemes have been built as different combinations of their components, i.e. feature pre-selection, elimination of missing values, automated feature selection, and a classifier. Two best performing schemes achieved the accuracy measured as Matthews correlation coefficient of 0.71 in test subset. These schemes involved W-LMT and W-SimpleLogistic classifiers. Significant differences have been observed between different classifiers and selected other components, depending on the dataset (validation or test). Discussed results may serve as guidelines to design a scheme to predict user satisfaction.
5.
Łukasz Radliński
Towards expert-based modeling of integrated software quality
In: Journal of Theoretical and Applied Computer Science, vol. 6, no. 2, pp. 13–26, 2012.
@article{Radlinski_jtacs2012,
title = {Towards expert-based modeling of integrated software quality},
author = {Łukasz Radliński},
url = {http://www.jtacs.org/archive/2012/2/2/JTACS_2012_02_02.pdf},
year = {2012},
date = {2012-01-01},
journal = {Journal of Theoretical and Applied Computer Science},
volume = {6},
number = {2},
pages = {13–26},
abstract = {This paper reports on a part of a project aimed at building an probabilistic model for integrated software quality simulation and prediction. This paper discusses results of the questionnaire survey focused on gathering expert knowledge about the factors influencing various features of software quality. Specifically, this analysis identifies project and process factors of software quality, investigates relationships between quality features and their sub-features as well as priorities for quality features. The survey has been performed among software engineering experts and projects managers. Obtained results will be used to calibrate that model for software quality simulation and prediction. These results also partially deliver a general overview on how software quality features are perceived by industry.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This paper reports on a part of a project aimed at building an probabilistic model for integrated software quality simulation and prediction. This paper discusses results of the questionnaire survey focused on gathering expert knowledge about the factors influencing various features of software quality. Specifically, this analysis identifies project and process factors of software quality, investigates relationships between quality features and their sub-features as well as priorities for quality features. The survey has been performed among software engineering experts and projects managers. Obtained results will be used to calibrate that model for software quality simulation and prediction. These results also partially deliver a general overview on how software quality features are perceived by industry.
6.
Łukasz Radliński; Jakub Swacha
C# or Java? – Analysis of Student Preferences
In: Studies and Proceedings of Polish Association for Knowledge Management, vol. 58, pp. 101–113, 2012.
@article{Radlinski_smpszw2012,
title = {C# or Java? – Analysis of Student Preferences},
author = {Łukasz Radliński and Jakub Swacha},
year = {2012},
date = {2012-01-01},
journal = {Studies and Proceedings of Polish Association for Knowledge Management},
volume = {58},
pages = {101–113},
abstract = {The aim of this paper is to determine the overall student preference to learning one of two programming languages: C# or Java. We performed a questionnaire survey among students of the subject field ‘computer science and econometrics’ at the University of Szczecin. All students were learning both C# and Java in the same semester and with the same course leader. We investigated obtained results for each of 13 individual questions. Then we aggregated them using Analytical Hierarchy Process (AHP) method to express students’ preferences quantitatively. The results show very strong dominance of student preferences to learning C# (73%) rather than Java (27%). In fact, according to the respondents, the only advantage of Java is its higher value in future professional career. We observed the strongest preference for C# by beginner part-time students, and the weakest preference for C# by beginner full-time students.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The aim of this paper is to determine the overall student preference to learning one of two programming languages: C# or Java. We performed a questionnaire survey among students of the subject field ‘computer science and econometrics’ at the University of Szczecin. All students were learning both C# and Java in the same semester and with the same course leader. We investigated obtained results for each of 13 individual questions. Then we aggregated them using Analytical Hierarchy Process (AHP) method to express students’ preferences quantitatively. The results show very strong dominance of student preferences to learning C# (73%) rather than Java (27%). In fact, according to the respondents, the only advantage of Java is its higher value in future professional career. We observed the strongest preference for C# by beginner part-time students, and the weakest preference for C# by beginner full-time students.
7.
Łukasz Radliński
On Generating Bayesian Nets from Small Local Qualitative Data for Software Development Effort and Quality Prediction
In: Metody Informatyki Stosowanej, vol. 27, no. 2, pp. 121–136, 2011.
@article{Radlinski_mis2011,
title = {On Generating Bayesian Nets from Small Local Qualitative Data for Software Development Effort and Quality Prediction},
author = {Łukasz Radliński},
year = {2011},
date = {2011-01-01},
journal = {Metody Informatyki Stosowanej},
volume = {27},
number = {2},
pages = {121–136},
abstract = {The aim of this paper is to investigate if it is possible to build accurate Bayesian net models for software development effort and quality prediction under two assumptions for model generation: (1) no expert knowledge is incorporated, (2) only small local qualitative data is used. Models generated in this study provide predictions with medium level of accuracy, yet still keeping the literature average. Thus, they can be used to make only rough estimations at the early software development stage. However, they can be a useful base for detailed models incorporating expert knowledge and tailored for individual needs.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The aim of this paper is to investigate if it is possible to build accurate Bayesian net models for software development effort and quality prediction under two assumptions for model generation: (1) no expert knowledge is incorporated, (2) only small local qualitative data is used. Models generated in this study provide predictions with medium level of accuracy, yet still keeping the literature average. Thus, they can be used to make only rough estimations at the early software development stage. However, they can be a useful base for detailed models incorporating expert knowledge and tailored for individual needs.
8.
Łukasz Radliński
A conceptual Bayesian net model for integrated software quality prediction
In: Annales UMCS, Informatica, vol. 11, no. 4, pp. 49–60, 2011, ISSN: 1732-1360.
@article{Radlinski_aumcsi2011,
title = {A conceptual Bayesian net model for integrated software quality prediction},
author = {Łukasz Radliński},
url = {http://versita.metapress.com/openurl.asp?genre=article&id=doi:10.2478/v10065-011-0032-5},
doi = {10.2478/v10065-011-0032-5},
issn = {1732-1360},
year = {2011},
date = {2011-01-01},
journal = {Annales UMCS, Informatica},
volume = {11},
number = {4},
pages = {49–60},
abstract = {Software quality can be described by a set of features, such as functionality, reliability, usability, efficiency, maintainability, portability and others. There are various models for software quality prediction developed in the past. Unfortunately, they typically focus on a single quality feature. The main goal of this study is to develop a predictive model that integrates several features of software quality, including relationships between them. This model is an expert-driven Bayesian net, which can be used in diverse analyses and simulations. The paper discusses model structure, behaviour, calibration and enhancement options, and possible use in fields other than software engineering.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Software quality can be described by a set of features, such as functionality, reliability, usability, efficiency, maintainability, portability and others. There are various models for software quality prediction developed in the past. Unfortunately, they typically focus on a single quality feature. The main goal of this study is to develop a predictive model that integrates several features of software quality, including relationships between them. This model is an expert-driven Bayesian net, which can be used in diverse analyses and simulations. The paper discusses model structure, behaviour, calibration and enhancement options, and possible use in fields other than software engineering.
9.
Łukasz Radliński
Factors of Software Quality - Analysis of Extended ISBSG Dataset
In: Foundations of Computing and Decision Studies, vol. 36, no. 3-4, pp. 293–313, 2011.
@article{Radlinski_fcds2011,
title = {Factors of Software Quality - Analysis of Extended ISBSG Dataset},
author = {Łukasz Radliński},
year = {2011},
date = {2011-01-01},
journal = {Foundations of Computing and Decision Studies},
volume = {36},
number = {3-4},
pages = {293–313},
abstract = {In this paper, we analyze the extended ISBSG dataset, which contains data on a wide range of software projects developed in various companies worldwide. The main aim of this paper is to identify important factors that influence software quality and to investigate the nature of these relationships. This analysis involves using various statistical techniques, both analytical and graphical. We provide a rating for each variable to express the strength of its relationship with software quality. Unlike earlier analyses, we focus on the business perspective and its relationships on software quality. Obtained results may be used do support decision making in software projects, specifically by demonstrating the impact of selected software development practices.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this paper, we analyze the extended ISBSG dataset, which contains data on a wide range of software projects developed in various companies worldwide. The main aim of this paper is to identify important factors that influence software quality and to investigate the nature of these relationships. This analysis involves using various statistical techniques, both analytical and graphical. We provide a rating for each variable to express the strength of its relationship with software quality. Unlike earlier analyses, we focus on the business perspective and its relationships on software quality. Obtained results may be used do support decision making in software projects, specifically by demonstrating the impact of selected software development practices.
10.
Łukasz Radliński
Techniki prognozowania nakładów projektowych i jakości oprogramowania w projektach IT
In: Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia Informatica, vol. 26, pp. 119–137, 2010.
@article{Radlinski_si26_2010,
title = {Techniki prognozowania nakładów projektowych i jakości oprogramowania w projektach IT},
author = {Łukasz Radliński},
year = {2010},
date = {2010-01-01},
journal = {Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia Informatica},
volume = {26},
pages = {119–137},
abstract = {The most important dimensions in software project estimation are: development effort and software quality. Several predictive models have been proposed for these dimensions. Although some of these models provide useful input for decision makers, most of them are inherently limited for industrial use. The aims of this study are to: (1) compare existing applications of methods and (2) select a best technique for building intelligent and practical models for development effort and software quality prediction. In recent years various techniques were used, which are based on statistics, machine learning, artificial intelligence and similar. Authors who used a single technique often report a success their studies, only sometimes additionally noticing threats in repeatability of their predictions in other environments. Other authors compare the accuracy of predictions obtained using different techniques. The main problem in these analyses is the lack of straightforward confirmation of the usefulness of specific techniques in building predictive models. When one author finds that one technique performs the best in their study, another author obtains the best predictions using a different technique. Bayesian nets (BNs) appear to be the best suited approach to build predictive intelligent and practical model. This paper summarizes some applications of BNs in modelling different aspects of software engineering and discusses proposed Productivity Model for analysing trade-offs between effort, size and software quality.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The most important dimensions in software project estimation are: development effort and software quality. Several predictive models have been proposed for these dimensions. Although some of these models provide useful input for decision makers, most of them are inherently limited for industrial use. The aims of this study are to: (1) compare existing applications of methods and (2) select a best technique for building intelligent and practical models for development effort and software quality prediction. In recent years various techniques were used, which are based on statistics, machine learning, artificial intelligence and similar. Authors who used a single technique often report a success their studies, only sometimes additionally noticing threats in repeatability of their predictions in other environments. Other authors compare the accuracy of predictions obtained using different techniques. The main problem in these analyses is the lack of straightforward confirmation of the usefulness of specific techniques in building predictive models. When one author finds that one technique performs the best in their study, another author obtains the best predictions using a different technique. Bayesian nets (BNs) appear to be the best suited approach to build predictive intelligent and practical model. This paper summarizes some applications of BNs in modelling different aspects of software engineering and discusses proposed Productivity Model for analysing trade-offs between effort, size and software quality.
11.
Łukasz Radliński
Integrating Data Sources in a Software Project Risk Assessment Model
In: Research Papers of Wrocław University of Economics, vol. 133, pp. 103–113, 2010.
@article{Radliński_rpwue2010,
title = {Integrating Data Sources in a Software Project Risk Assessment Model},
author = {Łukasz Radliński},
year = {2010},
date = {2010-01-01},
journal = {Research Papers of Wrocław University of Economics},
volume = {133},
pages = {103–113},
abstract = {This paper discusses integrated software project risk assessment model called the Productivity Model. There are two levels of integration involved with the model: (1) integrating effort, functionality and quality estimation and enabling trade-off analysis between these key project variables; (2) integrated knowledge base used to build the model made of various independent sources: expert knowledge, questionnaire survey results, results from statistical analysis of empirical data, and other reported data. Since the benefits of the first level of integration have been previously discussed this paper focuses on the second level by discussing the process of developing the Productivity Model using different data sources. Although this model reflects the software engineering area, such approach can be followed in building predictive models also in other fields.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This paper discusses integrated software project risk assessment model called the Productivity Model. There are two levels of integration involved with the model: (1) integrating effort, functionality and quality estimation and enabling trade-off analysis between these key project variables; (2) integrated knowledge base used to build the model made of various independent sources: expert knowledge, questionnaire survey results, results from statistical analysis of empirical data, and other reported data. Since the benefits of the first level of integration have been previously discussed this paper focuses on the second level by discussing the process of developing the Productivity Model using different data sources. Although this model reflects the software engineering area, such approach can be followed in building predictive models also in other fields.
12.
Łukasz Radliński
Ulepszone szacowanie ryzyka w przedsięwzięciach informatycznych z wykorzystaniem sieci Bayesa
In: Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia i Prace Wydziału Nauk Ekonomicznych i Zarządzania, vol. 20, pp. 147–161, 2010.
@article{Radlinski_spwneiz2010,
title = {Ulepszone szacowanie ryzyka w przedsięwzięciach informatycznych z wykorzystaniem sieci Bayesa},
author = {Łukasz Radliński},
year = {2010},
date = {2010-01-01},
journal = {Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia i Prace Wydziału Nauk Ekonomicznych i Zarządzania},
volume = {20},
pages = {147–161},
abstract = {Empirical software engineering models typically focus on predicting development effort or software quality but not both. Using Bayesian Nets (BNs) as causal models, researchers have recently attempted to build models that incorporate relationships between functionality, effort, software quality, and various process variables. The thesis analyses such models and, as part of a new validation study, identifies their strengths and weaknesses. A major weakness is their inability to incorporate prior local productivity and quality data, which limits their applicability in real software projects. The main hypothesis is that it is possible to build BN models that overcome these limitations without compromising their basic philosophy. In particular, the thesis shows we can build BNs that capture known trade-offs and can be tailored to individual company needs. The new model, called the productivity model, is developed by using the results of the new validation of the existing models, together with various other analyses. These include: the results of applying various statistical methods to identify relationships between a range of variables using publicly available data on software projects; analyses of other studies; expert knowledge. The new model is also calibrated using the results of an extensive questionnaire survey of experts in the area. The thesis also makes a number of other novel contributions to improved risk assessment using BNs, including a model which predicts the proportions of different types of defects likely to be left in software after testing and a learning model for predicting the number of defects found and fixed in successive testing iterations.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Empirical software engineering models typically focus on predicting development effort or software quality but not both. Using Bayesian Nets (BNs) as causal models, researchers have recently attempted to build models that incorporate relationships between functionality, effort, software quality, and various process variables. The thesis analyses such models and, as part of a new validation study, identifies their strengths and weaknesses. A major weakness is their inability to incorporate prior local productivity and quality data, which limits their applicability in real software projects. The main hypothesis is that it is possible to build BN models that overcome these limitations without compromising their basic philosophy. In particular, the thesis shows we can build BNs that capture known trade-offs and can be tailored to individual company needs. The new model, called the productivity model, is developed by using the results of the new validation of the existing models, together with various other analyses. These include: the results of applying various statistical methods to identify relationships between a range of variables using publicly available data on software projects; analyses of other studies; expert knowledge. The new model is also calibrated using the results of an extensive questionnaire survey of experts in the area. The thesis also makes a number of other novel contributions to improved risk assessment using BNs, including a model which predicts the proportions of different types of defects likely to be left in software after testing and a learning model for predicting the number of defects found and fixed in successive testing iterations.
13.
Łukasz Radliński
Przegląd sieci Bayesa do szacowania ryzyka w inżynierii oprogramowania
In: Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia Informatica, vol. 21, pp. 119–129, 2009.
@article{Radlinski_si21_2009,
title = {Przegląd sieci Bayesa do szacowania ryzyka w inżynierii oprogramowania},
author = {Łukasz Radliński},
year = {2009},
date = {2009-01-01},
journal = {Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia Informatica},
volume = {21},
pages = {119–129},
abstract = {The aim of this work is to compare selected Bayesian Nets for software project risk assessment. Previous studies have shown that Bayesian Nets have several advantages over other methods in creating models in the domain of software engineering. Thus, I did not analyze other models in this work. The results of analysis have shown that because of big differences between the models caused by different motivations for building them, direct comparison is difficult. Many of those models reflect only small part of software engineering. All of them suffer common weaknesses which are the reasons for my future work focused on creating models without those weaknesses.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The aim of this work is to compare selected Bayesian Nets for software project risk assessment. Previous studies have shown that Bayesian Nets have several advantages over other methods in creating models in the domain of software engineering. Thus, I did not analyze other models in this work. The results of analysis have shown that because of big differences between the models caused by different motivations for building them, direct comparison is difficult. Many of those models reflect only small part of software engineering. All of them suffer common weaknesses which are the reasons for my future work focused on creating models without those weaknesses.
14.
Łukasz Radliński
Przegląd publicznie dostępnych baz danych przedsięwzięć informatycznych
In: Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia Informatica, vol. 22, pp. 107–120, 2009.
@article{Radlinski_si22_2009,
title = {Przegląd publicznie dostępnych baz danych przedsięwzięć informatycznych},
author = {Łukasz Radliński},
year = {2009},
date = {2009-01-01},
journal = {Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia Informatica},
volume = {22},
pages = {107–120},
abstract = {A set of reliable empirical data is often required for a scientific research. For many years in the software engineering domain such datasets were usually not easily publicly available. However, re- cently some repositories have been established and opened for public access. This paper focuses on the analysis of databases of software projects. I have analyzed such databases grouped in four major re- positories: ISBSG, PROMISE, NASA and the databases based on Bugzilla. The contents are diverse among the databases: different parameters describing projects, different level of data granularity and different number of observations. Because of that there is no single database which could be used in each type of research analysis. Rather the aim of the research and the need for specific type of data determines which datad be used. A couple of possible types of analysis can be supported by these datasets: estimation of software size, effort and defects. A few of them allow the trade-off analysis between these factors. Some also contain data about the process and people quality. The comparison of the databases may be a useful tip when the choice of the database needs to be made.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
A set of reliable empirical data is often required for a scientific research. For many years in the software engineering domain such datasets were usually not easily publicly available. However, re- cently some repositories have been established and opened for public access. This paper focuses on the analysis of databases of software projects. I have analyzed such databases grouped in four major re- positories: ISBSG, PROMISE, NASA and the databases based on Bugzilla. The contents are diverse among the databases: different parameters describing projects, different level of data granularity and different number of observations. Because of that there is no single database which could be used in each type of research analysis. Rather the aim of the research and the need for specific type of data determines which datad be used. A couple of possible types of analysis can be supported by these datasets: estimation of software size, effort and defects. A few of them allow the trade-off analysis between these factors. Some also contain data about the process and people quality. The comparison of the databases may be a useful tip when the choice of the database needs to be made.
15.
Łukasz Radliński
Predicting Defect Types in Software Projects
In: Polish Journal of Environmental Studies, vol. 18, no. 3B, pp. 311–315, 2009.
@article{Radlinski_pjoes2009,
title = {Predicting Defect Types in Software Projects},
author = {Łukasz Radliński},
year = {2009},
date = {2009-01-01},
journal = {Polish Journal of Environmental Studies},
volume = {18},
number = {3B},
pages = {311–315},
abstract = {Predicting software defects has been one of the most demanding tasks for software engineering researchers and practitioners. The work in this area resulted in producing various defect prediction models. Their common weakness is that they typically treat all defects equally. However, software companies need to categorize defects found in their products to estimate user satisfaction and to prioritize which defects are supposed to be fixed first. Several classifications are supported by popular defect tracking systems, such as Bugzilla. The Defect Types Model introduced in this paper estimates proportions of defects of various types categorized by their severity for users. This model contains two groups of factors: controllable (process quality) and uncontrollable (features of software). It incorporates results from statistical analysis of the ISBSG project dataset adjusted by other reported results and expert knowledge. This model can be used either as a standalone model (predicting proportions of defects) or in combination with other defect prediction model (predicting number of defects of different types).},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Predicting software defects has been one of the most demanding tasks for software engineering researchers and practitioners. The work in this area resulted in producing various defect prediction models. Their common weakness is that they typically treat all defects equally. However, software companies need to categorize defects found in their products to estimate user satisfaction and to prioritize which defects are supposed to be fixed first. Several classifications are supported by popular defect tracking systems, such as Bugzilla. The Defect Types Model introduced in this paper estimates proportions of defects of various types categorized by their severity for users. This model contains two groups of factors: controllable (process quality) and uncontrollable (features of software). It incorporates results from statistical analysis of the ISBSG project dataset adjusted by other reported results and expert knowledge. This model can be used either as a standalone model (predicting proportions of defects) or in combination with other defect prediction model (predicting number of defects of different types).
16.
Łukasz Radliński; Norman Fenton; Martin Neil
A Learning Bayesian Net for Predicting Number of Software Defects Found in a Sequence of Testing
In: Polish Journal of Environmental Studies, vol. 17, no. 3B, pp. 359–364, 2008.
@article{Radlinski_pjoes2008,
title = {A Learning Bayesian Net for Predicting Number of Software Defects Found in a Sequence of Testing},
author = {Łukasz Radliński and Norman Fenton and Martin Neil},
year = {2008},
date = {2008-01-01},
journal = {Polish Journal of Environmental Studies},
volume = {17},
number = {3B},
pages = {359–364},
abstract = {We present the model for predicting the number of defects found in each testing iteration. The model assumes that the software is developed either according to the classical waterfall lifecycle or its variation, where testing takes place in a series of iterations after the coding of the whole software is finished. The input data to the initial model are the numbers of defects found in some first iterations of testing. Our model can also use as the input the estimated total number of residual defects given by other models. We partially validate the model using the publicly available datasets of software projects developed in cooperation with NASA: JM1, KC1, PC1, PC3, PC4. However, these datasets did not contain any information on testing effort and our results highlight the importance of recording such data for factors influencing testing effectiveness. We show how the model can be extended by adding other input variables such as effort, process and people quality and bias. Finally, we show the results of validating this extended model using a semi-randomly generated dataset.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
We present the model for predicting the number of defects found in each testing iteration. The model assumes that the software is developed either according to the classical waterfall lifecycle or its variation, where testing takes place in a series of iterations after the coding of the whole software is finished. The input data to the initial model are the numbers of defects found in some first iterations of testing. Our model can also use as the input the estimated total number of residual defects given by other models. We partially validate the model using the publicly available datasets of software projects developed in cooperation with NASA: JM1, KC1, PC1, PC3, PC4. However, these datasets did not contain any information on testing effort and our results highlight the importance of recording such data for factors influencing testing effectiveness. We show how the model can be extended by adding other input variables such as effort, process and people quality and bias. Finally, we show the results of validating this extended model using a semi-randomly generated dataset.
17.
Łukasz Radliński; Norman Fenton; Martin Neil; David Marquez
Modelling Prior Productivity and Defect Rates in a Causal Model for Software Project Risk Assessment
In: Polish Journal of Environmental Studies, vol. 16, no. 4A, pp. 256–260, 2007.
@article{Radlinski_pjoes2007,
title = {Modelling Prior Productivity and Defect Rates in a Causal Model for Software Project Risk Assessment},
author = {Łukasz Radliński and Norman Fenton and Martin Neil and David Marquez},
year = {2007},
date = {2007-01-01},
journal = {Polish Journal of Environmental Studies},
volume = {16},
number = {4A},
pages = {256–260},
abstract = {Estimating effort and quality of developed software has been a demanding task for project managers. Many models based on different approaches have been proposed to solve this problem. Most of them dealt only with one of either estimating effort or quality. Our aim is to develop a causal model capturing both of these and enabling trade-off analysis between the functionality, effort and quality. This model incorporates several quantitative and qualitative factors influencing the software development process and product. One of the key challenges is to be able to use some prior knowledge about the productivity and defect rates in the model. This knowledge can be obtained from the literature, a company’s data about past projects or assessed by an appropriate software project management expert. In this paper we present the results of an analysis which we performed using mainly ISBSG database of software projects. We also compared them with analyses available in the literature. These results are incorporated in our model for more accurate predictions especially when software companies do not have appropriate data about their past projects.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Estimating effort and quality of developed software has been a demanding task for project managers. Many models based on different approaches have been proposed to solve this problem. Most of them dealt only with one of either estimating effort or quality. Our aim is to develop a causal model capturing both of these and enabling trade-off analysis between the functionality, effort and quality. This model incorporates several quantitative and qualitative factors influencing the software development process and product. One of the key challenges is to be able to use some prior knowledge about the productivity and defect rates in the model. This knowledge can be obtained from the literature, a company’s data about past projects or assessed by an appropriate software project management expert. In this paper we present the results of an analysis which we performed using mainly ISBSG database of software projects. We also compared them with analyses available in the literature. These results are incorporated in our model for more accurate predictions especially when software companies do not have appropriate data about their past projects.
18.
Łukasz Radliński
Analiza porównawcza modeli jakości oprogramowania
In: Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia Informatica, vol. 19, pp. 131–150, 2006.
@article{Radlinski_si19_2006,
title = {Analiza porównawcza modeli jakości oprogramowania},
author = {Łukasz Radliński},
year = {2006},
date = {2006-01-01},
journal = {Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia Informatica},
volume = {19},
pages = {131–150},
abstract = {The software quality is one of the most important aspect of software. The aim of this work is to compare chosen software quality models. Several point of view on software quality have been discussed. Software quality models have been presented and compared according to specified criteria.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The software quality is one of the most important aspect of software. The aim of this work is to compare chosen software quality models. Several point of view on software quality have been discussed. Software quality models have been presented and compared according to specified criteria.
19.
Łukasz Radliński
Modelling Complex Nodes in Bayesian Nets for Software Project Risk Assessment
In: Polish Journal of Environmental Studies, vol. 15, no. 4C, pp. 149–152, 2006.
@article{Radlinski_pjoes2006,
title = {Modelling Complex Nodes in Bayesian Nets for Software Project Risk Assessment},
author = {Łukasz Radliński},
year = {2006},
date = {2006-01-01},
journal = {Polish Journal of Environmental Studies},
volume = {15},
number = {4C},
pages = {149–152},
abstract = {It is difficult to incorporate the nodes which are not directly observable into the Bayesian Nets. Using “soft evidence” only partially solves the problem. The real solutions are using either “indicators” or “causal nodes”. Each of them has several advantages and disadvantages so it is impossible to point out which one is the better in any case.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
It is difficult to incorporate the nodes which are not directly observable into the Bayesian Nets. Using “soft evidence” only partially solves the problem. The real solutions are using either “indicators” or “causal nodes”. Each of them has several advantages and disadvantages so it is impossible to point out which one is the better in any case.
20.
Łukasz Radliński
Software reliability estimation and prediction
In: Prace Naukowe Akademii Ekonomicznej we Wrocławiu, vol. 1044, pp. 82–90, 2004.
@article{Radlinski_ntiz2004,
title = {Software reliability estimation and prediction},
author = {Łukasz Radliński},
year = {2004},
date = {2004-01-01},
journal = {Prace Naukowe Akademii Ekonomicznej we Wrocławiu},
volume = {1044},
pages = {82–90},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
21.
Łukasz Radliński
Wybrane problemy zapewnienia jakości specyfikacji wymagań użytkowników
In: Prace Naukowe Akademii Ekonomicznej we Wrocławiu, vol. 896, pp. 69–77, 2003.
@article{Radlinski_ntiz2003,
title = {Wybrane problemy zapewnienia jakości specyfikacji wymagań użytkowników},
author = {Łukasz Radliński},
year = {2003},
date = {2003-01-01},
journal = {Prace Naukowe Akademii Ekonomicznej we Wrocławiu},
volume = {896},
pages = {69–77},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
22.
Łukasz Radliński
Pomiar jakości oprogramowania – możliwości i ograniczenia
In: Firma i Rynek, vol. 27-29, pp. 136–139, 2003.
@article{Radlinski_fir2003,
title = {Pomiar jakości oprogramowania – możliwości i ograniczenia},
author = {Łukasz Radliński},
year = {2003},
date = {2003-01-01},
journal = {Firma i Rynek},
volume = {27-29},
pages = {136–139},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
23.
Łukasz Radliński
Współczesne metody integracji w systemach informacyjnych zarządzania
In: Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia Informatica, vol. 17, 2003.
@article{Radlinski_si17_2003,
title = {Współczesne metody integracji w systemach informacyjnych zarządzania},
author = {Łukasz Radliński},
year = {2003},
date = {2003-01-01},
journal = {Zeszyty Naukowe Uniwersytetu Szczecińskiego, Seria: Studia Informatica},
volume = {17},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Book chapters
1.
Łukasz Radliński
How software development factors influence user satisfaction in meeting business objectives and requirements?
In: Madeyski, Lech; Ochodek, Mirosław (Ed.): Software Engineering from Research and Practice Perspectives, pp. 101–119, Nakom, Poznań-Warszawa, 2014.
@incollection{Radlinski_kkio2014_satisfaction,
title = {How software development factors influence user satisfaction in meeting business objectives and requirements?},
author = {Łukasz Radliński},
editor = {Lech Madeyski and Mirosław Ochodek},
year = {2014},
date = {2014-01-01},
booktitle = {Software Engineering from Research and Practice Perspectives},
pages = {101–119},
publisher = {Nakom},
address = {Poznań-Warszawa},
chapter = {6},
abstract = {User satisfaction is an useful measure of success of software development projects. The goal of this chapter is to analyze if and how individual factors describing software development pro-cess and product are related with selected features of user satisfaction. This chapter investigates two features of user satisfaction: meeting stated objectives (MSO) and meeting business requirements (MBR). Achieving such goal involved using visual techniques as well as a range of statistical and data mining techniques. For MSO there are more identified relationships and they are usually stronger than for MBR. Although there are some disagreements in relationships identified with different techniques, there is a common set of explanatory variables identified by most of techniques. Identified relationships can be used to build more complex simulation or predictive models.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
User satisfaction is an useful measure of success of software development projects. The goal of this chapter is to analyze if and how individual factors describing software development pro-cess and product are related with selected features of user satisfaction. This chapter investigates two features of user satisfaction: meeting stated objectives (MSO) and meeting business requirements (MBR). Achieving such goal involved using visual techniques as well as a range of statistical and data mining techniques. For MSO there are more identified relationships and they are usually stronger than for MBR. Although there are some disagreements in relationships identified with different techniques, there is a common set of explanatory variables identified by most of techniques. Identified relationships can be used to build more complex simulation or predictive models.
2.
Jakub Jurkiewicz; Piotr Kosiuczenko; Lech Madeyski; Mirosław Ochodek; Cezary Orłowski; Łukasz Radliński
Recent Polish achievements in Software Engineering
In: Madeyski, Lech; Ochodek, Mirosław (Ed.): Software Engineering from Research and Practice Perspectives, pp. 15–37, Nakom, Poznań-Warszawa, 2014.
@incollection{Jurkiewicz_kkio2014_advances,
title = {Recent Polish achievements in Software Engineering},
author = {Jakub Jurkiewicz and Piotr Kosiuczenko and Lech Madeyski and Mirosław Ochodek and Cezary Orłowski and Łukasz Radliński},
editor = {Lech Madeyski and Mirosław Ochodek},
year = {2014},
date = {2014-01-01},
booktitle = {Software Engineering from Research and Practice Perspectives},
pages = {15–37},
publisher = {Nakom},
address = {Poznań-Warszawa},
chapter = {1},
abstract = {Publications in top research journals (indexed by ISI) as well as citations are crucial in any research field to position the work and to build on the work of others. The objective of this chapter is twofold: to give an overview of the achievements of Polish research centers in the field of software engineering since 2010, and to present few recent contributions by researchers with Polish affiliations in ISI journals in the field of software engineering or closely related fields.},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
Publications in top research journals (indexed by ISI) as well as citations are crucial in any research field to position the work and to build on the work of others. The objective of this chapter is twofold: to give an overview of the achievements of Polish research centers in the field of software engineering since 2010, and to present few recent contributions by researchers with Polish affiliations in ISI journals in the field of software engineering or closely related fields.
3.
Łukasz Radliński
Empirical Analysis of the Impact of Requirements Engineering on Software Quality
In: Proc. International Working Conference on Requirements Engineering: Foundation for Software Quality, vol. 7195, pp. 232–238, Springer, Essen, 2012.
@inbook{Radlinski_refsq2012,
title = {Empirical Analysis of the Impact of Requirements Engineering on Software Quality},
author = {Łukasz Radliński},
url = {http://www.springerlink.com/content/e12w831642617740/},
doi = {10.1007/978-3-642-28714-5_21},
year = {2012},
date = {2012-01-01},
booktitle = {Proc. International Working Conference on Requirements Engineering: Foundation for Software Quality},
volume = {7195},
pages = {232–238},
publisher = {Springer},
address = {Essen},
series = {Lecture Notes in Computer Science},
abstract = {[Context & motivation] The process of requirements engineering affects software quality. However, empirical evaluation of this impact is required. [Question/problem] This paper aims to answer the following questions: (1) which factors related to requirements engineering affect software quality, (2) what is the nature of these relationships, and (3) how are soft quality features related to each other? [Principal ideas/results] To answer these questions we performed a quantitative and visual analysis using the extended ISBSG dataset. Obtained results cover a discussion on identified and unconfirmed relationships. [Contribution] The main contribution is an investigation of the relationships between factors of requirements engineering and software quality. Provided results can be used in further research and to guide industrial decision makers. The main limitation in generalizing the results is related to the high number of missing values in the dataset.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
[Context & motivation] The process of requirements engineering affects software quality. However, empirical evaluation of this impact is required. [Question/problem] This paper aims to answer the following questions: (1) which factors related to requirements engineering affect software quality, (2) what is the nature of these relationships, and (3) how are soft quality features related to each other? [Principal ideas/results] To answer these questions we performed a quantitative and visual analysis using the extended ISBSG dataset. Obtained results cover a discussion on identified and unconfirmed relationships. [Contribution] The main contribution is an investigation of the relationships between factors of requirements engineering and software quality. Provided results can be used in further research and to guide industrial decision makers. The main limitation in generalizing the results is related to the high number of missing values in the dataset.
4.
Thomas Schulz; Łukasz Radliński; Thomas Gorges; Wolfgang Rosenstiel
Software Process Model using Dynamic Bayesian Networks
In: Ramachandran, Muthu (Ed.): Knowledge Engineering for Software Development Life Cycles: Support Technologies and Applications, pp. 289–310, Information Science Reference, Hershey, 2011, ISBN: 978-1-60960-509-4.
@inbook{Schulz_igi2011,
title = {Software Process Model using Dynamic Bayesian Networks},
author = {Thomas Schulz and Łukasz Radliński and Thomas Gorges and Wolfgang Rosenstiel},
editor = {Muthu Ramachandran},
doi = {10.4018/978-1-60960-509-4.ch016},
isbn = {978-1-60960-509-4},
year = {2011},
date = {2011-01-01},
booktitle = {Knowledge Engineering for Software Development Life Cycles: Support Technologies and Applications},
pages = {289–310},
publisher = {Information Science Reference},
address = {Hershey},
abstract = {This chapter describes a methodology to support the management of large scale software projects in optimizing product correction effort versus initial development costs over time. The Software Process Model (SPM) represents the implementation of this approach on a level of detail explicitly developed to meet project manager’s demands. The underlying technique used in this approach is based on Dynamic Bayesian Networks (DBNs). The use of Bayesian Networks (BNs) enables the representation of causal relationships among process and product key performance indicators elicited either by data or expert knowledge. DBNs provide an insight into various aspects of SPM over time to assess current as well as predicting future model states. The objective of this chapter is to describe the practical approach to establish SPM as state of the art decision support in an industrial environment.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
This chapter describes a methodology to support the management of large scale software projects in optimizing product correction effort versus initial development costs over time. The Software Process Model (SPM) represents the implementation of this approach on a level of detail explicitly developed to meet project manager’s demands. The underlying technique used in this approach is based on Dynamic Bayesian Networks (DBNs). The use of Bayesian Networks (BNs) enables the representation of causal relationships among process and product key performance indicators elicited either by data or expert knowledge. DBNs provide an insight into various aspects of SPM over time to assess current as well as predicting future model states. The objective of this chapter is to describe the practical approach to establish SPM as state of the art decision support in an industrial environment.
5.
Łukasz Radliński
A Framework for Integrated Software Quality Prediction using Bayesian Nets
In: Proceedings of International Conference on Computational Science and Its Applications (ICCSA 2011), pp. 310–325, Springer, Santander, 2011.
@inbook{Radlinski_iccsa2011,
title = {A Framework for Integrated Software Quality Prediction using Bayesian Nets},
author = {Łukasz Radliński},
doi = {10.1007/978-3-642-21934-4_26},
year = {2011},
date = {2011-01-01},
booktitle = {Proceedings of International Conference on Computational Science and Its Applications (ICCSA 2011)},
pages = {310–325},
publisher = {Springer},
address = {Santander},
series = {Lecture Notes in Computer Science},
abstract = {The aim of this study is to develop a framework for integrated software quality prediction. This integration is reflected by a range of quality attributes incorporated in the model as well as relationships between these attributes. The model is formulated as a Bayesian net, a technique that has already been used in various software engineering studies. The framework enables to incorporate expert knowledge about the domain as well as related empirical data and encode them in the Bayesian net model. Such model may be used in decision support for software analytics and managers.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
The aim of this study is to develop a framework for integrated software quality prediction. This integration is reflected by a range of quality attributes incorporated in the model as well as relationships between these attributes. The model is formulated as a Bayesian net, a technique that has already been used in various software engineering studies. The framework enables to incorporate expert knowledge about the domain as well as related empirical data and encode them in the Bayesian net model. Such model may be used in decision support for software analytics and managers.
6.
Łukasz Radliński
Building Bayesian Nets for Software Defect Prediction - A Comparison of Manual, Semi- and Fully-Automated Schemes
In: Borzemski, Leszek; Grzech, Adam; Świątek, Jerzy; Wilimowska, Zofia (Ed.): Information Systems Architecture and Technology. New Developments in Web-Age Information Systems, pp. 321–335, Oficyna Wydawnicza Politechniki Wrocławskiej, Wrocław, 2010.
@inbook{Radlinski_isat2010,
title = {Building Bayesian Nets for Software Defect Prediction - A Comparison of Manual, Semi- and Fully-Automated Schemes},
author = {Łukasz Radliński},
editor = {Leszek Borzemski and Adam Grzech and Jerzy Świątek and Zofia Wilimowska},
year = {2010},
date = {2010-01-01},
booktitle = {Information Systems Architecture and Technology. New Developments in Web-Age Information Systems},
pages = {321–335},
publisher = {Oficyna Wydawnicza Politechniki Wrocławskiej},
address = {Wrocław},
abstract = {Various techniques have been used in software defect prediction. Bayesian nets (BNs) is one of such techniques. BNs have various advantages, especially important in an environment where no large dataset is available, as is typical for software defect prediction. BNs can be built using various schemes, depending on availability of domain knowledge and/or data: (1) automatically from the data, (2) manually by an expert or (3) using a mixture of expert knowledge and empirical data. This chapter discusses an experiment, which aim is to analyze and assess the accuracy of BNs for software defect prediction developed using these three schemes. An important assumption is to use a small set with data on past projects completed in a single company, with most of predictors describing development process quality. The results show that a model built by experts performs the most accurately to predict number of defects and defect rate; models generated using semi- and fully-automated schemes achieve significantly lower accuracy. Two main reasons of such result are: (1) low volume of data used to automatically generate the model structure and parameters and (2) the need to discretize numeric variables using wide intervals.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Various techniques have been used in software defect prediction. Bayesian nets (BNs) is one of such techniques. BNs have various advantages, especially important in an environment where no large dataset is available, as is typical for software defect prediction. BNs can be built using various schemes, depending on availability of domain knowledge and/or data: (1) automatically from the data, (2) manually by an expert or (3) using a mixture of expert knowledge and empirical data. This chapter discusses an experiment, which aim is to analyze and assess the accuracy of BNs for software defect prediction developed using these three schemes. An important assumption is to use a small set with data on past projects completed in a single company, with most of predictors describing development process quality. The results show that a model built by experts performs the most accurately to predict number of defects and defect rate; models generated using semi- and fully-automated schemes achieve significantly lower accuracy. Two main reasons of such result are: (1) low volume of data used to automatically generate the model structure and parameters and (2) the need to discretize numeric variables using wide intervals.
7.
Norman Fenton; Peter Hearty; Martin Neil; Łukasz Radliński
Software Project and Quality Modelling Using Bayesian Networks
In: Meziane, Farid; Vadera, Sunil (Ed.): Artificial Intelligence Applications for Improved Software Engineering Development: New Prospects, pp. 1–25, Information Science Reference, Hershey-New York, 2010.
@inbook{Fenton_igi2010,
title = {Software Project and Quality Modelling Using Bayesian Networks},
author = {Norman Fenton and Peter Hearty and Martin Neil and Łukasz Radliński},
editor = {Farid Meziane and Sunil Vadera},
doi = {10.4018/978-1-60566-758-4.ch001},
year = {2010},
date = {2010-01-01},
booktitle = {Artificial Intelligence Applications for Improved Software Engineering Development: New Prospects},
pages = {1–25},
publisher = {Information Science Reference},
address = {Hershey-New York},
abstract = {This chapter provides an introduction to the use of Bayesian Network (BN) models in Software Engineering. A short overview of the theory of BNs is included, together with an explanation of why BNs are ideally suited to dealing with the characteristics and shortcomings of typical software development environments. This theory is supplemented and illustrated using real world models that illustrate the advantages of BNs in dealing with uncertainty, causal reasoning and learning in the presence of limited data.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
This chapter provides an introduction to the use of Bayesian Network (BN) models in Software Engineering. A short overview of the theory of BNs is included, together with an explanation of why BNs are ideally suited to dealing with the characteristics and shortcomings of typical software development environments. This theory is supplemented and illustrated using real world models that illustrate the advantages of BNs in dealing with uncertainty, causal reasoning and learning in the presence of limited data.
8.
Łukasz Radliński; Norman Fenton
Causal Risk Framework for Software Projects
In: Wilimowska, Zofia; Borzemski, Leszek; Grzech, Adam; Świątek, Jerzy (Ed.): Information Systems Architecture and Technology. IT Technologies in Knowledge Oriented Management Process, pp. 49–59, Oficyna Wydawnicza Politechniki Wrocławskiej, Wrocław, Poland, 2009.
@inbook{Radlinski_isat2009,
title = {Causal Risk Framework for Software Projects},
author = {Łukasz Radliński and Norman Fenton},
editor = {Zofia Wilimowska and Leszek Borzemski and Adam Grzech and Jerzy Świątek},
year = {2009},
date = {2009-01-01},
booktitle = {Information Systems Architecture and Technology. IT Technologies in Knowledge Oriented Management Process},
pages = {49–59},
publisher = {Oficyna Wydawnicza Politechniki Wrocławskiej},
address = {Wrocław, Poland},
abstract = {A traditional approach to quantitative risk assessment is to calculate the "risk score" of an event as the product of its "probability" and "impact". But the "probability" and "impact" of an event cannot meaningfully be evaluated without incorporating factors that could prevent an event from occurring or to limit the negative consequences if it does occur. The "Risk score" is often interpreted as expected financial loss. Project managers are often interested in estimating other types of consequences such as: missing the deadline, need for additional staff, lower quality of delivered product etc. Bayesian nets (BNs) have the potential to extend traditionally perceived risk assessment by adding several features such as causality, uncertainty, mixture of empirical data with expert knowledge, and intuitiveness. In this paper we propose a causal risk framework (CRF) for software projects, which significantly extends traditional risk assessment. The CRF is based on risk categorisation depending on user perspective (role in the project). We use five components of risk: "trigger", "control", "risk event", "mitigant", "consequence". Analysis of existing BNs built for the software engineering area reveals that the potential use for most of them is limited because they are not directly compatible with CRF. We discuss the benefits of applying CRF to a Software Project Trade-off Model and directions for future research in applying CRF to software projects.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
A traditional approach to quantitative risk assessment is to calculate the "risk score" of an event as the product of its "probability" and "impact". But the "probability" and "impact" of an event cannot meaningfully be evaluated without incorporating factors that could prevent an event from occurring or to limit the negative consequences if it does occur. The "Risk score" is often interpreted as expected financial loss. Project managers are often interested in estimating other types of consequences such as: missing the deadline, need for additional staff, lower quality of delivered product etc. Bayesian nets (BNs) have the potential to extend traditionally perceived risk assessment by adding several features such as causality, uncertainty, mixture of empirical data with expert knowledge, and intuitiveness. In this paper we propose a causal risk framework (CRF) for software projects, which significantly extends traditional risk assessment. The CRF is based on risk categorisation depending on user perspective (role in the project). We use five components of risk: "trigger", "control", "risk event", "mitigant", "consequence". Analysis of existing BNs built for the software engineering area reveals that the potential use for most of them is limited because they are not directly compatible with CRF. We discuss the benefits of applying CRF to a Software Project Trade-off Model and directions for future research in applying CRF to software projects.
9.
Łukasz Radliński; Norman Fenton; David Marquez
Estimating Productivity and Defect Rates Based on Environmental Factors
In: Information Systems Architecture and Technology: Models of the Organisation's Risk Management, pp. 103–113, Oficyna Wydawnicza Politechniki Wrocławskiej, Wrocław, Poland, 2008.
@inbook{Radlinski_isat2008,
title = {Estimating Productivity and Defect Rates Based on Environmental Factors},
author = {Łukasz Radliński and Norman Fenton and David Marquez},
year = {2008},
date = {2008-01-01},
booktitle = {Information Systems Architecture and Technology: Models of the Organisation's Risk Management},
pages = {103–113},
publisher = {Oficyna Wydawnicza Politechniki Wrocławskiej},
address = {Wrocław, Poland},
abstract = {In previous work we have developed a causal model for software project risk assessment, called the Productivity Model. It enables trade-off analysis between key project factors: effort, size and quality. In this paper we build a model (which we call the PDR model) providing more informed prior estimates of productivity and defect rates as inputs to the Productivity Model. The PDR model has the Na"ıve Bayesian Classifier (NBC) structure. We built the PDR model using results from analysis of the ISBSG dataset which we adjusted by other known results and expert knowledge. The assumption for the PDR model was to use only the factors which describe the environment in which the software is developed such as: language type, development platform, methodology and CASE tool used. Therefore, we do not expect the model to be extremely accurate since it does not capture any process factors (such factors are accounted for in the Productivity Model). We still believe that, although PDR model has only limited value in isolation, it performs in a consistent way and may significantly change the estimates given by the Productivity Model.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
In previous work we have developed a causal model for software project risk assessment, called the Productivity Model. It enables trade-off analysis between key project factors: effort, size and quality. In this paper we build a model (which we call the PDR model) providing more informed prior estimates of productivity and defect rates as inputs to the Productivity Model. The PDR model has the Na"ıve Bayesian Classifier (NBC) structure. We built the PDR model using results from analysis of the ISBSG dataset which we adjusted by other known results and expert knowledge. The assumption for the PDR model was to use only the factors which describe the environment in which the software is developed such as: language type, development platform, methodology and CASE tool used. Therefore, we do not expect the model to be extremely accurate since it does not capture any process factors (such factors are accounted for in the Productivity Model). We still believe that, although PDR model has only limited value in isolation, it performs in a consistent way and may significantly change the estimates given by the Productivity Model.
10.
Łukasz Radliński; Norman Fenton; David Marquez; Peter Hearty
Empirical Analysis of Software Defect Types
In: Information Systems Architecture and Technology. Information Technology and Web Engineering: Models, Concepts & Challenges, pp. 223–231, Oficyna Wydawnicza Politechniki Wrocławskiej, Wrocław, Poland, 2007.
@inbook{Radlinski_isat2007,
title = {Empirical Analysis of Software Defect Types},
author = {Łukasz Radliński and Norman Fenton and David Marquez and Peter Hearty},
year = {2007},
date = {2007-01-01},
booktitle = {Information Systems Architecture and Technology. Information Technology and Web Engineering: Models, Concepts & Challenges},
pages = {223–231},
publisher = {Oficyna Wydawnicza Politechniki Wrocławskiej},
address = {Wrocław, Poland},
abstract = {There have been numerous models predicting software defect counts or probability that a given software part (e.g. module) is defective. The need for such models is unquestionable. However, in many cases software managers would be interested not just in number of defects which are likely to occur in software but also the number of defects by severity (such as extreme, major, minor). In this paper we present results of analysing data from the ISBSG database of software projects focusing on identification of factors influencing types of defects as well as the nature of these influences. We also demonstrate a Bayesian Net (BN) – a causal model that enables us to predict defect types using a combination of quantitative and qualitative factors. This model incorporates results of analysis based on several sources of publicly available data about the past projects, mainly ISBSG dataset and other literature. This model can be combined with a previously developed BN model that integrates resource and defect prediction.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
There have been numerous models predicting software defect counts or probability that a given software part (e.g. module) is defective. The need for such models is unquestionable. However, in many cases software managers would be interested not just in number of defects which are likely to occur in software but also the number of defects by severity (such as extreme, major, minor). In this paper we present results of analysing data from the ISBSG database of software projects focusing on identification of factors influencing types of defects as well as the nature of these influences. We also demonstrate a Bayesian Net (BN) – a causal model that enables us to predict defect types using a combination of quantitative and qualitative factors. This model incorporates results of analysis based on several sources of publicly available data about the past projects, mainly ISBSG dataset and other literature. This model can be combined with a previously developed BN model that integrates resource and defect prediction.
11.
Norman Fenton; Łukasz Radliński; Martin Neil
Improved Bayesian Networks for Software Project Risk Assessment Using Dynamic Discretisation
In: Software Engineering Techniques: Design for Quality, pp. 139–148, Springer, Boston, MA, 2006.
@inbook{Fenton_set2006,
title = {Improved Bayesian Networks for Software Project Risk Assessment Using Dynamic Discretisation},
author = {Norman Fenton and Łukasz Radliński and Martin Neil},
doi = {10.1007/978-0-387-39388-9_14},
year = {2006},
date = {2006-01-01},
booktitle = {Software Engineering Techniques: Design for Quality},
pages = {139–148},
publisher = {Springer},
address = {Boston, MA},
abstract = {It is possible to build useful models for software project risk assessment based on Bayesian networks. A number of such models have been published and used and they provide valuable predictions for decision-makers. However, the accuracy of the published models is limited due to the fact that they are based on crudely discretised numeric nodes. In traditional Bayesian network tools such discretisation was inevitable; modelers had to decide in advance how to split a numeric range into appropriate intervals taking account of the trade-off between model efficiency and accuracy. However, recent a recent breakthrough algorithm now makes dynamic discretisation practical. We apply this algorithm to existing software project risk models. We compare the accuracy of predictions and calculation time for models with and without dynamic discretisation nodes.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
It is possible to build useful models for software project risk assessment based on Bayesian networks. A number of such models have been published and used and they provide valuable predictions for decision-makers. However, the accuracy of the published models is limited due to the fact that they are based on crudely discretised numeric nodes. In traditional Bayesian network tools such discretisation was inevitable; modelers had to decide in advance how to split a numeric range into appropriate intervals taking account of the trade-off between model efficiency and accuracy. However, recent a recent breakthrough algorithm now makes dynamic discretisation practical. We apply this algorithm to existing software project risk models. We compare the accuracy of predictions and calculation time for models with and without dynamic discretisation nodes.
12.
Jakub Swacha; Łukasz Radliński; Marcin Bandosz
Metody udostępniania materiałów dydaktycznych poprzez Internet
In: Dydaktyka informatyki i informatyka w dydaktyce, Printshop, Szczecin, Poland, 2006.
@inbook{Swacha_dyd2006,
title = {Metody udostępniania materiałów dydaktycznych poprzez Internet},
author = {Jakub Swacha and Łukasz Radliński and Marcin Bandosz},
year = {2006},
date = {2006-01-01},
booktitle = {Dydaktyka informatyki i informatyka w dydaktyce},
publisher = {Printshop},
address = {Szczecin, Poland},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
13.
Edward Kolbusz; Jadwiga Cyparska; Jerzy Marcinkiewicz; Łukasz Radliński
Analiza systemu infomatycznego
In: Kolbusz, Edward; Olejniczak, Wojciech; Szyjewski, Zdzisław (Ed.): Inżynieria systemów informatycznych w e-gospodarce, Polskie Wydawnictwo Ekonomiczne, Warsaw, Poland, 2005.
@inbook{Kolbusz_ana2005,
title = {Analiza systemu infomatycznego},
author = {Edward Kolbusz and Jadwiga Cyparska and Jerzy Marcinkiewicz and Łukasz Radliński},
editor = {Edward Kolbusz and Wojciech Olejniczak and Zdzisław Szyjewski},
year = {2005},
date = {2005-01-01},
booktitle = {Inżynieria systemów informatycznych w e-gospodarce},
publisher = {Polskie Wydawnictwo Ekonomiczne},
address = {Warsaw, Poland},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
14.
Edward Kolbusz; Łukasz Radliński
Systemy informatyczne w e-gospodarce
In: Kolbusz, Edward; Olejniczak, Wojciech; Szyjewski, Zdzisław (Ed.): Inżynieria systemów informatycznych w e-gospodarce, Polskie Wydawnictwo Ekonomiczne, Warsaw, Poland, 2005.
@inbook{Kolbusz_is2005,
title = {Systemy informatyczne w e-gospodarce},
author = {Edward Kolbusz and Łukasz Radliński},
editor = {Edward Kolbusz and Wojciech Olejniczak and Zdzisław Szyjewski},
year = {2005},
date = {2005-01-01},
booktitle = {Inżynieria systemów informatycznych w e-gospodarce},
publisher = {Polskie Wydawnictwo Ekonomiczne},
address = {Warsaw, Poland},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
15.
Witold M. Paczkowski; Aleksandra Radlińska; Mateusz Radliński; Łukasz Radliński
Polioptymalizacyjna analiza kratowego przekrycia walcowego
In: Jankowski, K. (Ed.): Trendy naukowe młodzieży akademickiej. Nauki ścisłe, pp. 105–110, Wydawnictwo Akademii Podlaskiej, Siedlce, Poland, 2004.
@inbook{Paczkowski_trendy2004,
title = {Polioptymalizacyjna analiza kratowego przekrycia walcowego},
author = {Witold M. Paczkowski and Aleksandra Radlińska and Mateusz Radliński and Łukasz Radliński},
editor = {K. Jankowski},
year = {2004},
date = {2004-01-01},
booktitle = {Trendy naukowe młodzieży akademickiej. Nauki ścisłe},
pages = {105–110},
publisher = {Wydawnictwo Akademii Podlaskiej},
address = {Siedlce, Poland},
abstract = {The aim of this paper is to present a quasi-evolutional polyoptimization analysis of a barrel vault spatial truss. A cross section catalogue of bars that enables to minimize the objective functions has been searched.The calculations have been made in two separate cycles with a use of already existing system Optytruss and yet created program Katalog. The presented analysis clearly reveals an effectiveness of an evolutional polyoptimization use in a process of structures' designing. It significantly helps to reduce time of calculations and the results obtained in its following cycles are being noticeably improved.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
The aim of this paper is to present a quasi-evolutional polyoptimization analysis of a barrel vault spatial truss. A cross section catalogue of bars that enables to minimize the objective functions has been searched.The calculations have been made in two separate cycles with a use of already existing system Optytruss and yet created program Katalog. The presented analysis clearly reveals an effectiveness of an evolutional polyoptimization use in a process of structures' designing. It significantly helps to reduce time of calculations and the results obtained in its following cycles are being noticeably improved.
16.
Witold M. Paczkowski; Aleksandra Radlińska; Mateusz Radliński; Łukasz Radliński
Quasi-evolutional polyoptymization of a barrel vault spatial truss
In: Polioptymalizacja i komputerowe wspomaganie projektowania, pp. 144–151, Wydawnictwa Naukowo-Techniczne, Mielno, 2004.
@inbook{Paczkowski_wnt2004,
title = {Quasi-evolutional polyoptymization of a barrel vault spatial truss},
author = {Witold M. Paczkowski and Aleksandra Radlińska and Mateusz Radliński and Łukasz Radliński},
year = {2004},
date = {2004-01-01},
booktitle = {Polioptymalizacja i komputerowe wspomaganie projektowania},
pages = {144–151},
publisher = {Wydawnictwa Naukowo-Techniczne},
address = {Mielno},
abstract = {The aim of this paper is to present a quasi-evolutional polyoptimization analysis of a barrel vault spatial truss. A cross section catalogue of bars that enables to minimize the objective functions has been searched.The calculations have been made in two separate cycles with a use of already existing system Optytruss and yet created program Katalog. The presented analysis clearly reveals an effectiveness of an evolutional polyoptimization use in a process of structures' designing. It significantly helps to reduce time of calculations and the results obtained in its following cycles are being noticeably improved.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
The aim of this paper is to present a quasi-evolutional polyoptimization analysis of a barrel vault spatial truss. A cross section catalogue of bars that enables to minimize the objective functions has been searched.The calculations have been made in two separate cycles with a use of already existing system Optytruss and yet created program Katalog. The presented analysis clearly reveals an effectiveness of an evolutional polyoptimization use in a process of structures' designing. It significantly helps to reduce time of calculations and the results obtained in its following cycles are being noticeably improved.
17.
Łukasz Radliński
Poprawa procesów programowych w małych organizacjach programistycznych
In: Informatyka narzędziem współczesnego zarządzania, pp. 437–446, Wydawnictwo Polsko-Japońskiej Wyższej Szkoły Technik Komputerowych, Warsaw, Poland, 2004.
@inbook{Radlinski_pol_jap2004,
title = {Poprawa procesów programowych w małych organizacjach programistycznych},
author = {Łukasz Radliński},
year = {2004},
date = {2004-01-01},
booktitle = {Informatyka narzędziem współczesnego zarządzania},
pages = {437–446},
publisher = {Wydawnictwo Polsko-Japońskiej Wyższej Szkoły Technik Komputerowych},
address = {Warsaw, Poland},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
18.
Łukasz Radliński
Możliwości rozwoju zdalnego nauczania na szczeblu akademickim w Polsce
In: Dydaktyka informatyki ekonomicznej - kształcenie dla społeczeństwa informacyjnego, Wydawnictwo Akademii Ekonomicznej im. Oskara Langego, Wrocław, Poland, 2003.
@inbook{Radlinski_dinfo2003,
title = {Możliwości rozwoju zdalnego nauczania na szczeblu akademickim w Polsce},
author = {Łukasz Radliński},
year = {2003},
date = {2003-01-01},
booktitle = {Dydaktyka informatyki ekonomicznej - kształcenie dla społeczeństwa informacyjnego},
publisher = {Wydawnictwo Akademii Ekonomicznej im. Oskara Langego},
address = {Wrocław, Poland},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
19.
Marcin Bandosz; Łukasz Radliński; Jakub Swacha
Methods of usage of the Internet in process of distance education
In: Zeszyt Jubileuszowy z okazji 10-lecia Niemieckojęzycznego Potoku Studiów Informatyki Ekonomicznej na Uniwersytecie Szczecińskim, Wydawnictwo Naukowe Uniwersytetu Szczecińskiego, Szczecin, Poland, 2002.
@inbook{Bandosz_jubil2002,
title = {Methods of usage of the Internet in process of distance education},
author = {Marcin Bandosz and Łukasz Radliński and Jakub Swacha},
year = {2002},
date = {2002-01-01},
booktitle = {Zeszyt Jubileuszowy z okazji 10-lecia Niemieckojęzycznego Potoku Studiów Informatyki Ekonomicznej na Uniwersytecie Szczecińskim},
publisher = {Wydawnictwo Naukowe Uniwersytetu Szczecińskiego},
address = {Szczecin, Poland},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Conference proceedings
1.
Łukasz Radliński
The Trade-off Between Data Volume and Quality in Predicting User Satisfaction in Software Projects
In: 2024 50th Euromicro Conference on Software Engineering and Advanced Applications (SEAA), pp. 483–490, IEEE Computer Society, Paris, 2024.
@inproceedings{Radlinski2024,
title = {The Trade-off Between Data Volume and Quality in Predicting User Satisfaction in Software Projects},
author = {Łukasz Radliński},
url = {http://lukaszradlinski.com/wp-content/uploads/SEAA_2024_satisfaction_prediction___camera_ready.pdf},
doi = {10.1109/SEAA64295.2024.00080},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {2024 50th Euromicro Conference on Software Engineering and Advanced Applications (SEAA)},
pages = {483–490},
publisher = {IEEE Computer Society},
address = {Paris},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
2.
Łukasz Radliński
Predicting User Satisfaction in Software Projects using Machine Learning Techniques
In: Ali, Raian; Kaindl, Hermann; Maciaszek, Leszek (Ed.): Proceedings of the 15th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE, pp. 374–381, SciTePress, 2020, ISBN: 978-989-758-421-3.
@inproceedings{Radlinski_enase2020,
title = {Predicting User Satisfaction in Software Projects using Machine Learning Techniques},
author = {Łukasz Radliński},
editor = {Raian Ali and Hermann Kaindl and Leszek Maciaszek},
url = {http://www.scitepress.org/DigitalLibrary/Link.aspx?doi=10.5220/0009391803740381},
doi = {10.5220/0009391803740381},
isbn = {978-989-758-421-3},
year = {2020},
date = {2020-01-01},
booktitle = {Proceedings of the 15th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE},
pages = {374–381},
publisher = {SciTePress},
abstract = {User satisfaction is an important aspect of software quality. Factors of user satisfaction and its impact on project success were analysed in various studies. However, very few studies investigated the ability to predict user satisfaction. This paper presents results of such challenge. The analysis was performed with the ISBSG dataset of software projects. The target variable, satisfaction score, was defined as a sum of eight variables reflecting different aspects of user satisfaction. Twelve machine learning algorithms were used to build 40 predictive models. Each model was evaluated on 20 passes with a test subset. On average, a random forest model with missing data imputation by mode and mean achieved the best performance with the macro mean absolute error of 1.88. Four variables with the highest importance on predictions for this model are: survey respondent role, log(effort estimate), log(summary work effort), and proportion of major defects. On average 14 models performed worse than a simple baseline model. While best performing models deliver predictions with satisfactory accuracy, high variability of performance between different model variants was observed. Thus, a careful selection of model settings is required when attempting to use such model in practise.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
User satisfaction is an important aspect of software quality. Factors of user satisfaction and its impact on project success were analysed in various studies. However, very few studies investigated the ability to predict user satisfaction. This paper presents results of such challenge. The analysis was performed with the ISBSG dataset of software projects. The target variable, satisfaction score, was defined as a sum of eight variables reflecting different aspects of user satisfaction. Twelve machine learning algorithms were used to build 40 predictive models. Each model was evaluated on 20 passes with a test subset. On average, a random forest model with missing data imputation by mode and mean achieved the best performance with the macro mean absolute error of 1.88. Four variables with the highest importance on predictions for this model are: survey respondent role, log(effort estimate), log(summary work effort), and proportion of major defects. On average 14 models performed worse than a simple baseline model. While best performing models deliver predictions with satisfactory accuracy, high variability of performance between different model variants was observed. Thus, a careful selection of model settings is required when attempting to use such model in practise.
3.
Łukasz Radliński
An expert-driven Bayesian network model for simulating and predicting software quality
In: Malzahn, Dirk (Ed.): Proc. Fifth International Conference on Information, Process, and Knowledge Management, pp. 26–31, IARIA, Nice, France, 2013.
@inproceedings{Radlinski_eknow2013,
title = {An expert-driven Bayesian network model for simulating and predicting software quality},
author = {Łukasz Radliński},
editor = {Dirk Malzahn},
url = {http://www.thinkmind.org/index.php?view=article&articleid=eknow_2013_2_50_60175},
year = {2013},
date = {2013-01-01},
booktitle = {Proc. Fifth International Conference on Information, Process, and Knowledge Management},
pages = {26–31},
publisher = {IARIA},
address = {Nice, France},
abstract = {The main goal of this work is to build an expert-driven Bayesian network model for simulating and predicting software quality. In contrast with earlier models, this model represents software quality as a hierarchy of features and their sub-features where the features are interrelated with other. It contains a range of project and process factors that influence particular quality features. It has been pre-calibrated using results from the questionnaire survey performed among software engineers and managers in various software organizations. Managers in software projects can use such model to simulate and predict various aspects of software quality, typically at the early stage of project lifecycle. Proposed may become a central part of the future decision support system aimed to analyze, understand, manage and optimize a software development process.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The main goal of this work is to build an expert-driven Bayesian network model for simulating and predicting software quality. In contrast with earlier models, this model represents software quality as a hierarchy of features and their sub-features where the features are interrelated with other. It contains a range of project and process factors that influence particular quality features. It has been pre-calibrated using results from the questionnaire survey performed among software engineers and managers in various software organizations. Managers in software projects can use such model to simulate and predict various aspects of software quality, typically at the early stage of project lifecycle. Proposed may become a central part of the future decision support system aimed to analyze, understand, manage and optimize a software development process.
4.
Łukasz Radliński
Enhancing Bayesian Network Model for Integrated Software Quality Prediction
In: Mauri, Jaime Lloret; Lorenz, Pascal (Ed.): Proc. Fourth International Conference on Information, Process, and Knowledge Management, pp. 144–149, IARIA, Valencia, 2012.
@inproceedings{Radlinski_eknow2012,
title = {Enhancing Bayesian Network Model for Integrated Software Quality Prediction},
author = {Łukasz Radliński},
editor = {Jaime Lloret Mauri and Pascal Lorenz},
url = {http://www.thinkmind.org/index.php?view=article&articleid=eknow_2012_6_30_60108},
year = {2012},
date = {2012-01-01},
booktitle = {Proc. Fourth International Conference on Information, Process, and Knowledge Management},
pages = {144–149},
publisher = {IARIA},
address = {Valencia},
abstract = {A Bayesian network model for integrated software quality prediction, proposed in earlier study, has potential in supporting decision makers in software projects. However, it also has some disadvantages limiting its use. The aim of this paper is to overcome these limitations by enhancing the original model in three ways: (1) incorporating project factors, (2) adding subnets with detailed process factors, and (3) modeling integration of software components or sub-systems. These enhancements significantly improve the analytical usefulness of this predictive Bayesian network model.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A Bayesian network model for integrated software quality prediction, proposed in earlier study, has potential in supporting decision makers in software projects. However, it also has some disadvantages limiting its use. The aim of this paper is to overcome these limitations by enhancing the original model in three ways: (1) incorporating project factors, (2) adding subnets with detailed process factors, and (3) modeling integration of software components or sub-systems. These enhancements significantly improve the analytical usefulness of this predictive Bayesian network model.
5.
Emadoddin Livani; Łukasz Radliński; Elham Paikari; Guenther Ruhe
A Hybrid Method for Scenario-Based Effort Re-allocation in Software Projects
In: Proc. International Conference on Software Engineering and Applications, Dallas, 2011.
@inproceedings{Livani_sea2011,
title = {A Hybrid Method for Scenario-Based Effort Re-allocation in Software Projects},
author = {Emadoddin Livani and Łukasz Radliński and Elham Paikari and Guenther Ruhe},
url = {http://www.actapress.com/Abstract.aspx?paperId=453032},
doi = {10.2316/P.2011.758-048},
year = {2011},
date = {2011-01-01},
booktitle = {Proc. International Conference on Software Engineering and Applications},
address = {Dallas},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
6.
Łukasz Radliński
Software Development Effort and Quality Prediction Using Bayesian Nets and Small Local Qualitative Data
In: Proc. 22nd International Conference on Software Engineering and Knowledge Engineering, pp. 113–116, Knowledge Systems Institute, Redwood City, CA, 2010.
@inproceedings{Radlinski_seke2010,
title = {Software Development Effort and Quality Prediction Using Bayesian Nets and Small Local Qualitative Data},
author = {Łukasz Radliński},
year = {2010},
date = {2010-01-01},
booktitle = {Proc. 22nd International Conference on Software Engineering and Knowledge Engineering},
pages = {113–116},
publisher = {Knowledge Systems Institute},
address = {Redwood City, CA},
abstract = {Large and homogeneous datasets are typically required to predict software development effort and quality accurately. Also, many statistical methods can only be applied when meeting various constraints. This study focuses on developing Bayesian nets (BNs) automatically from a small local dataset. Predictive accuracy of generated BNs keeps the level of other published results but the procedure of building the models is simpler The accuracy can be improved by incorporating domain expert knowledge.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Large and homogeneous datasets are typically required to predict software development effort and quality accurately. Also, many statistical methods can only be applied when meeting various constraints. This study focuses on developing Bayesian nets (BNs) automatically from a small local dataset. Predictive accuracy of generated BNs keeps the level of other published results but the procedure of building the models is simpler The accuracy can be improved by incorporating domain expert knowledge.
7.
Thomas Schulz; Łukasz Radliński; Thomas Gorges; Wolfgang Rosenstiel
Defect Cost Flow Model - A Bayesian Network for Predicting Defect Correction Effort
In: Proc. 6th International Conference on Predictive Models in Software Engineering, ACM Press, Timisoara, 2010.
@inproceedings{Schulz_promise2010,
title = {Defect Cost Flow Model - A Bayesian Network for Predicting Defect Correction Effort},
author = {Thomas Schulz and Łukasz Radliński and Thomas Gorges and Wolfgang Rosenstiel},
year = {2010},
date = {2010-01-01},
booktitle = {Proc. 6th International Conference on Predictive Models in Software Engineering},
publisher = {ACM Press},
address = {Timisoara},
abstract = {Background. Software defect prediction has been one of the central topics of software engineering. Predicted defect counts have been used mainly to assess software quality and estimate the defect correction effort (DCE). However, in many cases these defect counts are not good indicators for DCE. Therefore, in this study DCE has been modeled from a different perspective. Defects originating from various development phases have different impact on the overall DCE, especially defects shifting from one phase to another. To reduce the DCE of a software product it is important to assess every development phase along with its specific characteristics and focus on the shift of defects over phases. Aims. The aim of this paper is to build a model for effort prediction at different development stages. Our model is mainly focused on a dynamic DCE changing from one development phase to another. It reflects the increasing cost of correcting defects which are introduced in early, but found in later development phases. Research Method. The modeling technique used in this study is a Bayesian network which, among many others, has three important capabilities: reflecting causal relationships, combining expert knowledge with empirical data and incorporating uncertainty. The procedure of model development contains a set of iterations including the following steps: problem analysis, data analysis, model enhancement with simulation runs and model validation. Results. The developed Defect Cost Flow Model (DCFM) reflects the widely used V-model, an international standard for developing information technology systems. It has been pre-calibrated with empirical data from past projects developed at Robert Bosch GmbH. The analysis of evaluation scenarios confirms that DCFM correctly incorporates known qualitative and quantitative relationships. Because of its causal structure it can be used intuitively by end-users. Conclusion. Typical cost benefit optimization strategies regarding the optimal effort spent on quality measures tend to optimize locally, e.g. every development phase is optimized separately in its own domain. In contrast to that, the DCFM demonstrates that even cost intensive quality measures pay off when the overall DCE of specific features is considered.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Background. Software defect prediction has been one of the central topics of software engineering. Predicted defect counts have been used mainly to assess software quality and estimate the defect correction effort (DCE). However, in many cases these defect counts are not good indicators for DCE. Therefore, in this study DCE has been modeled from a different perspective. Defects originating from various development phases have different impact on the overall DCE, especially defects shifting from one phase to another. To reduce the DCE of a software product it is important to assess every development phase along with its specific characteristics and focus on the shift of defects over phases. Aims. The aim of this paper is to build a model for effort prediction at different development stages. Our model is mainly focused on a dynamic DCE changing from one development phase to another. It reflects the increasing cost of correcting defects which are introduced in early, but found in later development phases. Research Method. The modeling technique used in this study is a Bayesian network which, among many others, has three important capabilities: reflecting causal relationships, combining expert knowledge with empirical data and incorporating uncertainty. The procedure of model development contains a set of iterations including the following steps: problem analysis, data analysis, model enhancement with simulation runs and model validation. Results. The developed Defect Cost Flow Model (DCFM) reflects the widely used V-model, an international standard for developing information technology systems. It has been pre-calibrated with empirical data from past projects developed at Robert Bosch GmbH. The analysis of evaluation scenarios confirms that DCFM correctly incorporates known qualitative and quantitative relationships. Because of its causal structure it can be used intuitively by end-users. Conclusion. Typical cost benefit optimization strategies regarding the optimal effort spent on quality measures tend to optimize locally, e.g. every development phase is optimized separately in its own domain. In contrast to that, the DCFM demonstrates that even cost intensive quality measures pay off when the overall DCE of specific features is considered.
8.
Milijana Fineman; Norman Fenton; Lukasz Radlinski
Modelling Project Trade-Off Using Bayesian Networks
In: Proc. International Conference on Computational Intelligence and Software Engineering, pp. 1–4, Wuhan, 2009, ISBN: 978-1-4244-4507-3.
@inproceedings{Fineman_cise2009,
title = {Modelling Project Trade-Off Using Bayesian Networks},
author = {Milijana Fineman and Norman Fenton and Lukasz Radlinski},
url = {http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=5364789},
isbn = {978-1-4244-4507-3},
year = {2009},
date = {2009-12-01},
booktitle = {Proc. International Conference on Computational Intelligence and Software Engineering},
pages = {1–4},
address = {Wuhan},
abstract = {This paper considers trade-offs that may be made during project among time, cost and quality. Most projects are unique one-of-a-kind activities. Hence, the project manager may find it extremely difficult to stay within the time-cost-quality triangle. Different project trade-off preferences exist in different industries. Based on this we propose set of axioms in project risk management which reflect relationships between key project variables. The conceptual model we develop satisfies this basic set of axioms. We use Bayesian networks (BNs) to model project trade-off. The purpose is to provide a quantitative model for trade-off analysis in project risk management.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper considers trade-offs that may be made during project among time, cost and quality. Most projects are unique one-of-a-kind activities. Hence, the project manager may find it extremely difficult to stay within the time-cost-quality triangle. Different project trade-off preferences exist in different industries. Based on this we propose set of axioms in project risk management which reflect relationships between key project variables. The conceptual model we develop satisfies this basic set of axioms. We use Bayesian networks (BNs) to model project trade-off. The purpose is to provide a quantitative model for trade-off analysis in project risk management.
9.
Norman Fenton; Martin Neil; William Marsh; Peter Hearty; Łukasz Radliński; Paul Krause
Project Data Incorporating Qualitative Factors for Improved Software Defect Prediction
In: Proceedings of the Third international Workshop on Predictor Models in Software Engineering. International Conference on Software Engineering, pp. 2, IEEE Computer Society, Washington, DC, 2007.
@inproceedings{Fenton_promise2007,
title = {Project Data Incorporating Qualitative Factors for Improved Software Defect Prediction},
author = {Norman Fenton and Martin Neil and William Marsh and Peter Hearty and Łukasz Radliński and Paul Krause},
doi = {10.1109/PROMISE.2007.11},
year = {2007},
date = {2007-01-01},
booktitle = {Proceedings of the Third international Workshop on Predictor Models in Software Engineering. International Conference on Software Engineering},
pages = {2},
publisher = {IEEE Computer Society},
address = {Washington, DC},
abstract = {To make accurate predictions of attributes like defects found in complex software projects we need a rich set of process factors. We have developed a causal model that includes such process factors, both quantitative and qualitative. The factors in the model were identified as part of a major collaborative project. A challenge for such a model is getting the data needed to validate it. We present a dataset, elicited from 31 completed software projects in the consumer electronics industry, which we used for validation. The data were gathered using a questionnaire distributed to managers of recent projects. The dataset will be of interest to other researchers evaluating models with similar aims. We make both the dataset and causal model available for research use.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
To make accurate predictions of attributes like defects found in complex software projects we need a rich set of process factors. We have developed a causal model that includes such process factors, both quantitative and qualitative. The factors in the model were identified as part of a major collaborative project. A challenge for such a model is getting the data needed to validate it. We present a dataset, elicited from 31 completed software projects in the consumer electronics industry, which we used for validation. The data were gathered using a questionnaire distributed to managers of recent projects. The dataset will be of interest to other researchers evaluating models with similar aims. We make both the dataset and causal model available for research use.
10.
Łukasz Radliński; Norman Fenton; Martin Neil; David Marqez
Improved Decision-Making for Software Managers Using Bayesian Networks
In: Proceedings of the 11th IASTED International Conference Software Engineering and Applications, pp. 13–19, IASTED, Cambridge, MA, 2007.
@inproceedings{Radlinski_sea2007,
title = {Improved Decision-Making for Software Managers Using Bayesian Networks},
author = {Łukasz Radliński and Norman Fenton and Martin Neil and David Marqez},
url = {http://portal.acm.org/citation.cfm?id=1647640},
year = {2007},
date = {2007-01-01},
booktitle = {Proceedings of the 11th IASTED International Conference Software Engineering and Applications},
pages = {13–19},
publisher = {IASTED},
address = {Cambridge, MA},
abstract = {Although there have been many models for predicting quality and resources in software development, they do not provide enough decision-support for software managers. It has been argued that models based on Bayesian Nets (BNs) address this problem. Unlike previous BN models, the new model (called the Productivity Model) presented here takes account of both up-to-date generic empirical data on software productivity and quality and, where relevant, local information and choice of units of measurement. The model provides comprehensive support for ‘trade-off’ analysis to help decision-making.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Although there have been many models for predicting quality and resources in software development, they do not provide enough decision-support for software managers. It has been argued that models based on Bayesian Nets (BNs) address this problem. Unlike previous BN models, the new model (called the Productivity Model) presented here takes account of both up-to-date generic empirical data on software productivity and quality and, where relevant, local information and choice of units of measurement. The model provides comprehensive support for ‘trade-off’ analysis to help decision-making.
11.
Aleksander Badower; Aleksandra Radlińska; Mateusz Radliński; Łukasz Radliński
Stochastyczny charakter przekrojów prętów katalogu. Ocena bezpieczeństwa konstrukcji
In: Materiały XXII Ogólnopolskiej Konferencji Polioptymalizacja i komputerowe wspomaganie projektowania. Streszczenia, Koszalin-Sarbinowo, 2004.
@inproceedings{Badower_poliopt2004,
title = {Stochastyczny charakter przekrojów prętów katalogu. Ocena bezpieczeństwa konstrukcji},
author = {Aleksander Badower and Aleksandra Radlińska and Mateusz Radliński and Łukasz Radliński},
year = {2004},
date = {2004-01-01},
booktitle = {Materiały XXII Ogólnopolskiej Konferencji Polioptymalizacja i komputerowe wspomaganie projektowania. Streszczenia},
address = {Koszalin-Sarbinowo},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
12.
Władysław Hoffmann; Karolina Banaś; Łukasz Radliński
Tendencje rozwoju biometrycznych metod identyfikacji osób
In: Materiały konferencyjne – Multimedialne i Sieciowe Systemy Informacyjne, Wrocław, Poland, 2002.
@inproceedings{Hoffmann_missi2002,
title = {Tendencje rozwoju biometrycznych metod identyfikacji osób},
author = {Władysław Hoffmann and Karolina Banaś and Łukasz Radliński},
year = {2002},
date = {2002-01-01},
booktitle = {Materiały konferencyjne – Multimedialne i Sieciowe Systemy Informacyjne},
address = {Wrocław, Poland},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
13.
Jakub Swacha; Marcin Bandosz; Łukasz Radliński
Zaawansowane multimedia na stronach WWW
In: Materiały konferencyjne – Multimedialne i Sieciowe Systemy Informacyjne, Wrocław, Poland, 2002.
@inproceedings{Swacha_missi2002,
title = {Zaawansowane multimedia na stronach WWW},
author = {Jakub Swacha and Marcin Bandosz and Łukasz Radliński},
year = {2002},
date = {2002-01-01},
booktitle = {Materiały konferencyjne – Multimedialne i Sieciowe Systemy Informacyjne},
address = {Wrocław, Poland},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Other
1.
Lukasz Radlinski
Improved Software Project Risk Assessment Using Bayesian Nets
Queen Mary, University of London, 2008.
@phdthesis{Radlinski_phd2008,
title = {Improved Software Project Risk Assessment Using Bayesian Nets},
author = {Lukasz Radlinski},
year = {2008},
date = {2008-01-01},
school = {Queen Mary, University of London},
abstract = {Empirical software engineering models typically focus on predicting development effort or software quality but not both. Using Bayesian Nets (BNs) as causal models, researchers have recently attempted to build models that incorporate relationships between functionality, effort, software quality, and various process variables. This thesis analyses such models and, as part of a new validation study, identifies their strengths and weaknesses. A major weakness is their inability to incorporate prior local productivity and quality data, which limits their applicability in real software projects. The main hypothesis is that it is possible to build BN models that overcome these limitations without compromising their basic philosophy. In particular, the thesis shows we can build BNs that capture known trade-offs and can be tailored to individual company needs. The new model, called the Productivity Model, is developed by using the results of the new validation of the existing model, together with various other analyses. These include: the results of applying various statistical methods to identify relationships between a range of variables using publicly available data on software projects; analyses of other studies; expert knowledge. The new model is also calibrated using the results of an extensive questionnaire survey of experts in the area. The thesis also makes a number of other novel contributions to improved risk assessment using BNs, including: - A model which predicts the proportions of different types of defects likely to be left in software after testing. The model uses the results of statistical analysis on the past software projects. It can be combined with other defect prediction models to predict the number of residual defects of different categories. - A learning model for predicting the number of defects found and fixed in successive testing iterations.},
keywords = {},
pubstate = {published},
tppubtype = {phdthesis}
}
Empirical software engineering models typically focus on predicting development effort or software quality but not both. Using Bayesian Nets (BNs) as causal models, researchers have recently attempted to build models that incorporate relationships between functionality, effort, software quality, and various process variables. This thesis analyses such models and, as part of a new validation study, identifies their strengths and weaknesses. A major weakness is their inability to incorporate prior local productivity and quality data, which limits their applicability in real software projects. The main hypothesis is that it is possible to build BN models that overcome these limitations without compromising their basic philosophy. In particular, the thesis shows we can build BNs that capture known trade-offs and can be tailored to individual company needs. The new model, called the Productivity Model, is developed by using the results of the new validation of the existing model, together with various other analyses. These include: the results of applying various statistical methods to identify relationships between a range of variables using publicly available data on software projects; analyses of other studies; expert knowledge. The new model is also calibrated using the results of an extensive questionnaire survey of experts in the area. The thesis also makes a number of other novel contributions to improved risk assessment using BNs, including: - A model which predicts the proportions of different types of defects likely to be left in software after testing. The model uses the results of statistical analysis on the past software projects. It can be combined with other defect prediction models to predict the number of residual defects of different categories. - A learning model for predicting the number of defects found and fixed in successive testing iterations.