ZFZhenlei Fu

Zhenlei Fu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Zhenlei Fu has a total of 7 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Software Fault Localization Based on SALSA Algorithm

    by Xin Fan, Zuxiong Shen, Zhenlei Fu, Yun Ge

    In software development, debugging is the most tedious and time-consuming phase. Therefore, various automated fault localization techniques have been proposed to assist debugging. Among existing fault localization techniques, Spectrum-Based Fault Localization (SBFL) is one of the most extensively researched methods. Traditional SBFL techniques rely solely on the coverage of program execution for fault localization, which means they neglect the interactions between program entities and fault propagation paths during the execution of the program, resulting in a tie problem that reduces the accuracy of fault localization. To solve the above problem, this paper proposes SA-SBFL, a fault localization method based on the SALSA (Random Method for Link Structure Analysis) algorithm. First, a link graph of program entities is constructed, which includes interactions between program entities and fault propagation paths. Then, the suspicion values obtained from traditional SBFL methods are used as the initial weights of the link graph. Finally, the random walk model is employed to simulate the propagation of faults among program entities, analyze the importance of program entities in the fault propagation process, and obtain a ranking list of suspicious program entities. The experiments in this paper demonstrate that the SA-SBFL method significantly outperforms general SBFL methods. For instance, in the Defects4J dataset, the SA-SBFL technique outperforms traditional SBFL in terms of fault localization accuracy, with a 47% improvement in the Top-1 metric and a 10% increase in the Top-5 metric, and it also showed an average improvement of 19% in the EXAM metric.

  • Open Access

    Two-Phase Software Fault Localization Based on Relational Graph Convolutional Neural Networks

    by Xin Fan, Jian Shu, Zuxiong Shen, Yun Ge, Zhenlei Fu

    Spectrum-based fault localization (SBFL) generates a ranked list of suspicious elements by using the program execution spectrum, but the excessive number of elements ranked in parallel results in low localization accuracy. Most researchers consider intra-class dependencies to improve localization accuracy. However, some studies show that inter-class method call type faults account for more than 20%, which means such methods still have certain limitations. To solve the above problems, this paper proposes a two-phase software fault localization based on relational graph convolutional neural networks (Two-RGCNFL). Firstly, in Phase 1, the method call dependence graph (MCDG) of the program is constructed, the intra-class and inter-class dependencies in MCDG are extracted by using the relational graph convolutional neural network, and the classifier is used to identify the faulty methods. Then, the GraphSMOTE algorithm is improved to alleviate the impact of class imbalance on classification accuracy. Aiming at the problem of parallel ranking of element suspicious values in traditional SBFL technology, in Phase 2, Doc2Vec is used to learn static features, while spectrum information serves as dynamic features. A RankNet model based on siamese multi-layer perceptron is constructed to score and rank statements in the faulty method. This work conducts experiments on 5 real projects of Defects4J benchmark. Experimental results show that, compared with the traditional SBFL technique and two baseline methods, our approach improves the Top-1 accuracy by 262.86%, 29.59% and 53.01%, respectively, which verifies the effectiveness of Two-RGCNFL. Furthermore, this work verifies the importance of inter-class dependencies through ablation experiments.