Cross Version Defect Prediction with Class Dependency Embeddings

TL;DR

This paper introduces a novel approach for cross version defect prediction by leveraging class dependency network embeddings combined with static code metrics, showing significant performance improvements.

Contribution

The work proposes using network embedding techniques on class dependency networks with alignment methods for defect prediction, outperforming traditional metrics-based models.

Findings

Meta-model with embeddings improves AUC by 4.7%

Network embeddings effectively capture structural information for defect prediction

Alignment techniques enhance cross-version embedding comparisons

Abstract

Software Defect Prediction aims at predicting which software modules are the most probable to contain defects. The idea behind this approach is to save time during the development process by helping find bugs early. Defect Prediction models are based on historical data. Specifically, one can use data collected from past software distributions, or Versions, of the same target application under analysis. Defect Prediction based on past versions is called Cross Version Defect Prediction (CVDP). Traditionally, Static Code Metrics are used to predict defects. In this work, we use the Class Dependency Network (CDN) as another predictor for defects, combined with static code metrics. CDN data contains structural information about the target application being analyzed. Usually, CDN data is analyzed using different handcrafted network measures, like Social Network metrics. Our approach uses network embedding techniques to leverage CDN information without having to build the metrics manually. In order to use the embeddings between versions, we incorporate different embedding alignment techniques. To evaluate our approach, we performed experiments on 24 software release pairs and compared it against several benchmark methods. In these experiments, we analyzed the performance of two different graph embedding techniques, three anchor selection approaches, and two alignment techniques. We also built a meta-model based on two different embeddings and achieved a statistically significant improvement in AUC of 4.7% (p < 0.002) over the baseline method.