We present a methodology for Bayesian analysis of software quality. We cast our research in the broader context of constructing a causal framework that can include process, product, and other diverse sources of information regarding fault introduction during the software development process. In this paper, we discuss the aspect of relating internal product metrics to external quality metrics. Specifically, we build a Bayesian network (BN) model to relate object-oriented software metrics to software fault content and fault proneness. Assuming that the relationship can be described as a generalized linear model, we derive parametric functional forms for the target node conditional distributions in the BN. These functional forms are shown to be able to represent linear, Poisson, and binomial logistic regression. The models are empirically evaluated using a public domain data set from a software subsystem. The results show that our approach produces statistically significant estimations and that our overall modeling method performs no worse than existing techniques.

Empirical Analysis of Software Fault Content and Fault Proneness Using Bayesian Methods

High-assurance and complex mission-critical software systems are heavily dependent on reliability of their underlying software applications. An early software fault prediction is a proven technique in achieving high software reliability. Prediction models based on software metrics can predict number of faults in software modules. Timely predictions of such models can be used to direct cost-effective quality enhancement efforts to modules that are likely to have a high number of faults. We evaluate the predictive performance of six commonly used fault prediction techniques: CART-LS (least squares), CART-LAD (least absolute deviation), S-PLUS, multiple linear regression, artificial neural networks, and case-based reasoning. The case study consists of software metrics collected over four releases of a very large telecommunications system. Performance metrics, average absolute and average relative errors, are utilized to gauge the accuracy of different prediction models. Models were built using both, original software metrics (RAW) and their principle components (PCA). Two-way ANOVA randomized-complete block design models with two blocking variables are designed with average absolute and average relative errors as response variables. System release and the model type (RAW or PCA) form the blocking variables and the prediction technique is treated as a factor. Using multiple-pairwise comparisons, the performance order of prediction models is determined. We observe that for both average absolute and average relative errors, the CART-LAD model performs the best while the S-PLUS model is ranked sixth.

Fault Prediction Modeling for Software Quality Estimation: Comparing Commonly Used Techniques

Complex high-assurance software systems depend highly on reliability of their underlying software applications. Early identification of high-risk modules can assist in directing quality enhancement efforts to modules that are likely to have a high number of faults. Regression tree models are simple and effective as software quality prediction models, and timely predictions from such models can be used to achieve high software reliability. This paper presents a case study from our comprehensive evaluation (with several large case studies) of currently available regression tree algorithms for software fault prediction. These are, CART-LS (least squares), S-PLUS, and CART-LAD (least absolute deviation). The case study presented comprises of software design metrics collected from a large network telecommunications system consisting of almost 13 million lines of code. Tree models using design metrics are built to predict the number of faults in modules. The algorithms are also compared based on the structure and complexity of their tree models. Performance metrics, average absolute and average relative errors are used to evaluate fault prediction accuracy.

Tree-based software quality estimation models for fault prediction

A software product line is a unified representation of a set of conceptually similar software systems that share many common features and satisfy the requirements of a particular domain. Within the context of software product lines, feature models are tree-like structures that are widely used for modeling and representing the inherent commonality and variability of software product lines. Given the fact that many different software systems can be spawned from a single software product line, it can be anticipated that a low-quality design can ripple through to many spawned software systems. Therefore, the need for early indicators of external quality attributes is recognized in order to avoid the implications of defective and low-quality design during the late stages of production. In this paper, we propose a set of structural metrics for software product line feature models and theoretically validate them using valid measurement-theoretic principles. Further, we investigate through controlled experimentation whether these structural metrics can be good predictors (early indicators) of the three main subcharacteristics of maintainability: analyzability, changeability, and understandability. More specifically, a four-step analysis is conducted: (1) investigating whether feature model structural metrics are correlated with feature model maintainability through the employment of classical statistical correlation techniques; (2) understanding how well each of the structural metrics can serve as discriminatory references for maintainability; (3) identifying the sufficient set of structural metrics for evaluating each of the subcharacteristics of maintainability; and (4) evaluating how well different prediction models based on the proposed structural metrics can perform in indicating the maintainability of a feature model. Results obtained from the controlled experiment support the idea that useful prediction models can be built for the purpose of evaluating feature model maintainability using early structural metrics. Some of the structural metrics show significant correlation with the subjective perception of the subjects about the maintainability of the feature models.

/pdf/assessing-the-maintainability-of-software-product-line-1kvgl0kocf.pdf

Assessing the maintainability of software product line feature models using structural metrics

Software defect prediction (SDP), which classifies software modules into defect-prone and not-defect-prone categories, provides an effective way to maintain high quality software systems. Most existing SDP models attempt to attain lower classification error rates other than lower misclassification costs. However, in many real-world applications, misclassifying defect-prone modules as not-defect-prone ones usually leads to higher costs than misclassifying not-defect-prone modules as defect-prone ones. In this paper, we first propose a new two-stage cost-sensitive learning (TSCS) method for SDP, by utilizing cost information not only in the classification stage but also in the feature selection stage. Then, specifically for the feature selection stage, we develop three novel cost-sensitive feature selection algorithms, namely, Cost-Sensitive Variance Score (CSVS), Cost-Sensitive Laplacian Score (CSLS), and Cost-Sensitive Constraint Score (CSCS), by incorporating cost information into traditional feature selection algorithms. The proposed methods are evaluated on seven real data sets from NASA projects. Experimental results suggest that our TSCS method achieves better performance in software defect prediction compared to existing single-stage cost-sensitive classifiers. Also, our experiments show that the proposed cost-sensitive feature selection methods outperform traditional cost-blind feature selection methods, validating the efficacy of using cost information in the feature selection stage.

Two-Stage Cost-Sensitive Learning for Software Defect Prediction

The objective in the construction of models of software quality is to use measures that may be obtained relatively early in the software development life cycle to provide reasonable initial estimates of the quality of an evolving software system. Measures of software quality and software complexity to be used in this modeling process exhibit systematic departures of the normality assumptions of regression modeling. Two new estimation procedures are introduced, and their performances in the modeling of software quality from software complexity in terms of the predictive quality and the quality of fit are compared with those of the more traditional least squares and least absolute value estimation techniques. The two new estimation techniques did produce regression models with better quality of fit and predictive quality when applied to data obtained from two software development projects. >

B.B. Bhattacharya

Papers

Predictive modeling techniques of software quality from software measures