Article
Computer Science, Software Engineering
Weifeng Pan, Hua Ming, Zijiang Yang, Tian Wang
Summary: In this comment, the authors criticize the top-core approach proposed by Qu et al. for improving the performance of bug prediction models. They point out three shortcomings in Qu et al.'s paper, and provide alternative solutions to address these issues. The experiments show that Qu et al.'s approach does not significantly outperform the state-of-the-art method, but replacing CDN with an improved CDN does improve the performance.
IEEE TRANSACTIONS ON SOFTWARE ENGINEERING
(2022)
Article
Computer Science, Software Engineering
Yu Qu, Qinghua Zheng, Jianlei Chi, Yangxu Jin, Ancheng He, Di Cui, Hengshan Zhang, Ting Liu
Summary: Complex Network theory and graph algorithms have been effective in predicting software bugs, with k-core decomposition being used to identify key classes. A new perspective on analyzing software bug distribution is presented using k-core decomposition on Class Dependency Networks. The proposed top-core equation significantly improves bug prediction models' performances and helps identify real bugs more quickly and easily.
IEEE TRANSACTIONS ON SOFTWARE ENGINEERING
(2021)
Article
Mathematics, Applied
Xin Du, Tian Wang, Liuhai Wang, Weifeng Pan, Chunlai Chai, Xinxin Xu, Bo Jiang, Jiale Wang
Summary: This article introduces an improved effort-aware bug prediction approach, which uses a weighted directed class dependency network to describe software structure and combines the coreness of each class with its relative risk to measure bug risk.
Article
Computer Science, Artificial Intelligence
Chao Wei, Junying Zhang, Xiguo Yuan, Zongzhen He, Guojun Liu, Jinhui Wu
Summary: This study proposes a novel method for predicting translation initiation sites in mRNA sequences based on a hybrid dependency network and deep learning framework. By explicitly modeling label dependencies among coding regions and between coding regions and translation initiation sites, this method achieves excellent prediction performance on benchmark gene datasets, surpassing existing state-of-the-art methods.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Ziye Zhu, Hanghang Tong, Yu Wang, Yun Li
Summary: This paper proposes a graph-based neural model BLoco for automated bug localization, which decomposes bug reports into bug clues and designs a code hierarchical network structure and a multilayer graph neural network to capture program behaviors and the relationship between bug reports and source code files.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Software Engineering
Yu Qu, Heng Yin
Summary: Research on bug prediction techniques utilizing network embedding has shown significant improvement in performance, with newly proposed algorithms such as ProNE showcasing the best results. Combining embedded vectors with traditional software engineering metrics has proven to be highly effective in enhancing bug prediction models.
EMPIRICAL SOFTWARE ENGINEERING
(2021)
Article
Computer Science, Hardware & Architecture
Sen Fang, You-shuai Tan, Tao Zhang, Zhou Xu, Hui Liu
Summary: This article proposes a novel approach using graph convolutional networks and weighted loss function for prioritizing bug reports, addressing manual priority assignment and data imbalance challenges. Experimental results show that the method outperforms baseline approaches across multiple open-source projects.
IEEE TRANSACTIONS ON RELIABILITY
(2021)
Article
Computer Science, Artificial Intelligence
Wei Zheng, JingYuan Cheng, Xiaoxue Wu, Ruiyang Sun, Xiaolong Wang, Xiaobing Sun
Summary: To improve the effectiveness of software security bug report (SBR) prediction, this study enhances supervised machine learning-based SBR prediction with software security domain knowledge. By establishing entity relationships and constructing knowledge graphs, the domain knowledge-guided approach achieved a 52% improvement in prediction effectiveness, according to experimental evaluation on 5 open-source SBR datasets.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Software Engineering
Wentao Wang, Faryn Dumont, Nan Niu, Glen Horton
Summary: Cyber attacks targeting software applications have a significant impact on our daily life. This article presents a novel approach to integrate the interdependency among high-level security requirements and use automated requirements tracing methods to identify product-level security requirements and their dependencies. Experimental results show significant recall improvements at 81 percent, demonstrating the value of this approach in connecting requirements engineering with security testing.
IEEE TRANSACTIONS ON SOFTWARE ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Haiyan Wu, Chaogeng Huang, Shengchun Deng
Summary: Aspect-based sentiment analysis (ABSA) aims to extract sentiment-target pairs in review sentences. Previous methods based on recurrent neural networks (RNNs) struggle with accurately capturing sentiment pairs. Recent research incorporates dependency information into structured models, achieving better results, while ignoring domain knowledge related to entities in the comments. This paper proposes a Knowledge-aware Dependency Graph Network (KDGN) that incorporates domain knowledge, dependency labels, and syntax path, showing significant improvement on the ABSA task.
INFORMATION FUSION
(2023)
Article
Computer Science, Information Systems
Hadi Jahanshahi, Mucahit Cevik
Summary: This paper proposes S-DABT, a bug triaging method that utilizes integer programming and machine learning techniques to assign bugs to suitable developers. By considering textual data, bug fixing costs, and bug dependencies, as well as the schedule of developers, S-DABT achieves a fair distribution of tasks and efficient utilization of developers' time, leading to reduced bug fixing time.
INFORMATION AND SOFTWARE TECHNOLOGY
(2022)
Article
Statistics & Probability
Yubai Yuan, Annie Qu
Summary: Link prediction, an important problem in network analysis, is addressed in this study. The proposed tensor-based joint network embedding method captures the dependency between pairwise and multi-way links and improves link prediction accuracy.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2023)
Article
Computer Science, Artificial Intelligence
Zixu Wang, Weiyuan Tong, Peng Li, Guixin Ye, Hao Chen, Xiaoqing Gong, Zhanyong Tang
Summary: Software defect prediction is crucial for identifying potentially defective modules early in the software development lifecycle. Cross-version defect prediction utilizes labeled defect data from prior versions to predict defects in current versions. Existing machine learning-based methods fail to model long and deep dependencies and have low detection efficiency for large-scale software projects.
COMPLEX & INTELLIGENT SYSTEMS
(2023)
Article
Computer Science, Information Systems
R. Siva, S. Kaliraj, B. Hariharan, N. Premkumar
Summary: In the software maintenance and development process, early bug detection is crucial for enhancing software efficiency, reliability, quality, and cost. This article proposes an optimized long short-term memory model for efficient bug prediction, which includes preprocessing, feature selection, and bug detection stages.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Chunying Zhou, Peng He, Cheng Zeng, Ju Ma
Summary: This paper proposes a method that combines semantic and structural information for software defect prediction, achieving improved performance.
INFORMATION AND SOFTWARE TECHNOLOGY
(2022)