Developing Software Bug Prediction Models Using Various Software Metrics as the Bug Indicators
نویسندگان
چکیده
The bug prediction effectiveness reasonably contributes towards enhancing quality of software. Bug indicators contribute significantly in determining the bug prediction approaches and help in achieving software reliability. Various comparative research studies have indicated that Depth of Inheritance (DIT), Weighted Method per Class (WMC), Coupling between Objects (CBO) and Lines of Code (LoC) have significantly established themselves as reliable bug indicators for comprehensive bug predictions. The researchers have carried out a quantitative research and have developed prediction models using above bug indicators as models input and have applied these models on open source projects (Camel and Ant). During this research, the results demonstrates that there is significant correlation between size oriented metrics (bug indicators) such as DIT, WMC, CBO, LoC and bugs. Overall, DIT takes dominance in achieving better impact on predicting bugs than WMC, CBO and LoC. The outcomes of the present research study would be of significance to software quality practitioners worldwide and would help them in prioritizing the efforts involved in bug prediction. Keywords—Bug Prediction; DIT; WMC; CBO; LoC; SRGM
منابع مشابه
Impact of restricted forward greedy feature selection technique on bug prediction
Several change metrics and source code metrics have been introduced and proved to be effective in bug prediction. Researchers performed comparative studies of bug prediction models built using the individual metrics as well as combination of these metrics. In this paper, we investigate the impact of feature selection in bug prediction models by analyzing the misclassification rates of these mod...
متن کاملOn the Non-Generalizability in Bug Prediction
Bug prediction is a technique used to estimate the most bug-prone entities in software systems. Bug prediction approaches vary in many design options, such as dependent variables, independent variables, and machine learning models. Choosing the right combination of design options to build an e↵ective bug predictor is hard. Previous studies do not consider this complexity and draw conclusions ba...
متن کاملAgainst the Mainstream in Bug Prediction
Bug prediction is a technique used to estimate the most bug-prone entities in software systems. Bug prediction approaches vary in many design options, such as dependent variables, independent variables, and machine learning models. Choosing the right combination of design options to build an effective bug predictor is hard. Previous studies do not consider this complexity and draw conclusions b...
متن کاملReasons for Non-Applicability of Software Entropy Metrics for Bug Prediction in Android
Abstract—Software Entropy Metrics for bug prediction have been validated on various software systems by different researchers. In our previous research, we have validated that Software Entropy Metrics calculated for Mozilla subsystem’s predict the future bugs reasonably well. In this study, the Software Entropy metrics are calculated for a subsystem of Android and it is noticed that these metri...
متن کاملEvaluating the impact of software metrics on defects prediction. Part 2
Software metrics are used as indicators of the quality of the developed software. Metrics can be collected from any software part such as: code, design, or requirements. In this paper, we evaluated several examples of design coupling metrics. Analysis and experiments follow hereinafter to demonstrate the use and value of those metrics. This is the second part for a paper we published in Compute...
متن کامل