Doctoral Dissertations
Date of Award
5-2024
Degree Type
Dissertation
Degree Name
Doctor of Philosophy
Major
Industrial Engineering
Major Professor
Anahita Khojandi
Committee Members
James Ostrowski, Kamesh Madduri, Bing Yao
Abstract
Efficiently identifying and resolving software defects is essential for producing high quality software. Early and accurate prediction of these defects plays a pivotal role in maintaining software quality. This dissertation focuses on advancing software defect prediction methodologies and practical applications by incorporating graph-based learning techniques and generative adversarial-based anomaly detection techniques. First, we present a novel approach to software defect prediction by introducing a graph-based defect ratio (GDR). This innovative metric leverages the intricate graph structure that captures the interdependencies among developers, commits, and repositories, offering a promising alternative to standard traditional features. This study highlights the potential for graph-based feature engineering as an alternative or complement to the use of conventional features in defect prediction models, offering a more efficient approach to enhance software quality. Second, we apply a new Generative Adversarial Network-based technique for detecting anomalies (GAN-AN). We tailor this technique to identify software defects early in the software development process. Our technique, which detects defects as anomalies, is compared against three established anomaly detection methods: autoencoder, isolation forest, and one-class support vector machine (OCSVM). Our investigation reveals that while GAN-AN surpasses existing state-of-the-art methods, the effectiveness of models can differ significantly. This variation in performance metrics implies that the choice of the most suitable model could depend on the particular dataset and the specific objectives of defect detection. Finally, we explore the efficiency of just-in-time software defect prediction (JIT-SDP) using a novel tri-partite graph structure and Relational Graph Convolutional Networks iv (R-GCNs). Our detailed examination of the connections among developers, commits, and repositories deepen our insight into the dynamics of software defects. The study demonstrates that the performance of advanced computational models in defect prediction is closely tied to the specific characteristics of the data and the context of the problem. This research contributes to the ongoing dialogue in software defect prediction, underscoring the importance of an approach that judiciously combines sophisticated computational techniques with more traditional, proven methods.
Recommended Citation
Soni, Aradhana, "Graph-Based and Anomaly Detection Learning Models for Just-in-Time Defect Prediction. " PhD diss., University of Tennessee, 2024.
https://trace.tennessee.edu/utk_graddiss/10168