No description available.

The present study is concerned with developing some statistical models to evaluate and analyze software reliability. We have developed the analytical structure of the logistic model to be used for testing and evaluating the reliability of a software package. The proposed model has been shown to be useful in the testing and debugging stages of the developmental process of a software package. It is important that prior to releasing a software package to marketing that we have achieved a target reliability with an acceptable degree of confidence. The proposed model has been evaluated and compared with several existing statistical models that are commonly used. Real software failure data was used for the comparison of the proposed logistic model with the others. The proposed model gives better results or it is equally effective. The logistic model was also used to model the mean time between failure of software packages. Real failure data was used to illustrate the usefulness of the proposed statistical procedures. Using the logistic model to characterize software failures we proceed to develop Bayesian analysis of the subject model. This modeling was based on two different difference equations whose parameters were estimated with Bayesian regressions subject to specific prior and mean square loss function.

As software becomes more important to society, the number, age, and complexity of systems grow. Software organizations require continuous process improvement to maintain the reliability, security, and quality of these software systems. Software organizations can utilize data from manual fault classification to meet their process improvement needs, but organizations lack the expertise or resources to implement them correctly. This dissertation addresses the need for the automation of software fault classification. Validation results show that automated fault classification, as implemented in the MiSFIT tool, can group faults of similar nature. The resulting classifications result in good agreement for common software faults with no manual effort. To evaluate the method and tool, I develop and apply an extended change taxonomy to classify the source code changes that repaired software faults from an open source project. MiSFIT clusters the faults based on the changes. I manually inspect a random sample of faults from each cluster to validate the results. The automatically classified faults are used to analyze the evolution of a software application over seven major releases. The contributions of this dissertation are an extended change taxonomy for software fault analysis, a method to cluster faults by the syntax of the repair, empirical evidence that fault distribution varies according to the purpose of the module, and the identification of project-specific trends from the analysis of the changes.

Les logiciels défectueux ont des conséquences coûteuses. Les développeurs de logiciels doivent identifier et réparer les composants défectueux dans leurs logiciels avant de les publier. De même, les utilisateurs doivent évaluer la qualité du logiciel avant son adoption. Cependant, la nature abstraite et les multiples dimensions de la qualité des logiciels entravent les organisations de mesurer leur qualités. Les métriques de qualité logicielle peuvent être utilisées comme proxies de la qualité du logiciel. Cependant, il est nécessaire de disposer d'une métrique de processus logiciel spécifique qui peut garantir des performances de prédiction de défaut meilleures et cohérentes, et cela dans de différents contextes. Cette recherche avait pour objectif de déterminer un prédicteur de défauts logiciels qui présente la meilleure performance de prédiction, nécessite moins d'efforts pour la détection et a un coût minimum de mauvaise classification des composants défectueux. En outre, l'étude inclut une analyse de l'effet de la combinaison de prédicteurs sur la performance d'un modèles de prédiction de défauts logiciels. Les données expérimentales proviennent de quatre projets OSS. La régression logistique et la régression linéaire ont été utilisées pour prédire les défauts. Les métriques Change Burst ont enregistré les valeurs les plus élevées pour les mesures de performance numérique, avaient les probabilités de détection de défaut les plus élevées et le plus faible coût de mauvaise classification des composants. / Faulty software have expensive consequences. To mitigate these consequences, software developers have to identify and fix faulty software components before releasing their products. Similarly, users have to gauge the delivered quality of software before adopting it. However, the abstract nature and multiple dimensions of software quality impede organizations from measuring software quality. Software quality metrics can be used as proxies of software quality. There is need for a software process metric that can guarantee consistent superior fault prediction performances across different contexts. This research sought to determine a predictor for software faults that exhibits the best prediction performance, requires least effort to detect software faults, and has a minimum cost of misclassifying components. It also investigated the effect of combining predictors on performance of software fault prediction models. Experimental data was derived from four OSS projects. Logistic Regression was used to predict bug status while Linear Regression was used to predict number of bugs per file. Models built with Change Burst metrics registered overall better performance relative to those built with Change, Code Churn, Developer Networks and Source Code software metrics. Change Burst metrics recorded the highest values for numerical performance measures, exhibited the highest fault detection probabilities and had the least cost of mis-classification of components. The study found out that Change Burst metrics could effectively predict software faults.