Data Mining for Predictors of Software Quality

Abstract

Knowledge discovery in data bases" (KDD) for software engineering is a process for finding useful information in the large volumes of data that are a byproduct of software development, such as data bases for configuration management and for problem reporting. This paper presents guidelines for extracting innovative process metrics from these commonly available data bases. This paper also adapts the Classification And Regression Trees algorithm, CART, to the KDD process for software engineering data. To our knowledge, this algorithm has not been used previously for empirical software quality modeling. In particular, we present an innovative way to control the balance between misclassification rates. A KDD case study of a very large legacy telecommunications software system found that variables derived from source code, configuration management transactions, and problem reporting transactions can be useful predictors of software quality. The KDD process discovered that for this software development environment, out of forty software attributes, only a few of the predictor variables were significant. This resulted in a model that predicts whether modules are likely to have faults discovered by customers. Software developers need such predictions early in development to target software enhancement techniques to the modules that need improvement the most. Int. J. Soft. Eng. Knowl. Eng. 1999.09:547-563. Downloaded from www.worldscientific.com by MONASH UNIVERSITY on 02/02/15. For personal use only.

Understanding the impact of code and process metrics on post-release defects

“…They also argued that change data is a better predictor of defects than code metrics in general. Studies by Arisholm and Briand [21] and Khoshgoftaar et al [22] also reported that prior changes are a good predictor of defects in a file. Hassan [3] used the complexity of a code change to predict defects.…”

Strong dependencies between software components

“…Component dependencies can be used to compute relevant quality measures of software repositories, for instance to identify particularly fragile components [7,13,15]. It is well known that small-world networks are resilient to random failures but particularly weak in the presence of attacks, due to the existence of highly connected hub nodes [2].…”

A method for detecting faulty code violating implicit coding rules

“…And, this method is also useful in decreasing the number of bugs made by maintainers who are not aware of implicit coding rules. In addition, this method it very powerful in practical use because it can directly detect the line number of faulty code area with detailed instructions, while previous methods for predicting fault-prone modules using software metrics only say which modules are faulty [4,7].…”