An extensive evaluation of ensemble techniques for software change prediction

Abstract

Predicting the areas of the source code having a higher likelihood to change in the future represents an important activity to allow developers to plan preventive maintenance operations.For this reason, several change prediction models have been proposed. Moreover, research community demonstrated how different classifiers impact on the performance of devised models as well as classifiers tend to perform similarly even though they are able to correctly predict the change proneness of different code elements, possibly indicating the presence of some complementarity among them. In this paper, we deeper investigated whether the use of ensemble approaches, ie, machine learning techniques able to combine multiple classifiers, can improve the performances of change prediction models. Specifically, we built three change prediction models based on different predictors, ie, product-, process-metrics-, and developer-related factors, comparing the performances of four ensemble techniques (ie, Boosting, Random Forest, Bagging, and Voting) with those of standard machine learning classifiers (ie, Logistic Regression, Naive Bayes, Simple Logistic, and Multilayer Perceptron). The study was conducted on 33releases of 10 open-source systems, and the results showed how ensemble methods and in particular Random Forest provide a significant improvement of more than 10% in terms of F measure. Indeed, the statistical analyses conducted confirm the superiority of this ensemble technique. Moreover, the model built using developer-related factors performed better than the other models that exploit product and process metrics and achieves an overall median of F measure around 77%.

Simplified abstract syntax tree based semantic features learning for software change prediction

“…4 They aim to predict the modules that need to be extensively modified in the future. Other researchers 5,8,10 count the number of times each module is changed and consider the modules that exceed the median number of changed times among all the modules of the system as change-prone. We found that more existing works adopt the latter approach to define the change-prone label, that is, the number of times the code changes.…”

“…It has been confirmed by many researchers for its correlation with changeproneness. 3,10,11 In addition, we follow the work of Yan et al 9 to select four Li and Henry indicators for measuring object-oriented software. The detailed explanation is presented in Table 4.…”

“…8 Developer metrics reflect the complexity of the development process and researchers hold that complex changes consume more maintenance resources. With these metrics, machine learning classifiers can be leveraged to build change prediction models, like logistic regression, 4,5,8,9 random forest, 10 and support vector machine. 4,11 However, these methods ignore the semantic information of the code.…”

Towards Pattern-Based Change Verification Framework for Cloud-Enabled Healthcare Component-Based

“…In this unit, we discuss different types of T.V. [53], [54] for our experiment i.e. internal validity threats (IVT), external validity threats (EVT), conclusion validity threats (CVT) and construct validity threats (CsVT) and how to resolve these limitations.…”

A critical review on blockchain assessment initiatives: A technology evolution viewpoint

“…One of the inherent characteristics of software is change. Given that software systems and their circumstances evolve with time, change is inevitable in the software arena 3 . Moreover, changes in software are key to correct existing defects and to incorporate improvements according to varying requirements or changes in the environment 4 …”