2020/01/01 by Ritu Kapur, Balwinder Sodhi, Kapur, Ritu +5
Computer Science · #Advanced Malware Detection Techniques #FOS: Computer and information sciences #Software Engineering (cs.SE) #Software Engineering Research #Software System Performance and Reliability #Topic Modeling
paper · pdf · doi:10.48550/arxiv.2104.14265
openalex publication_date 2021/04/29 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Code reviews are one of the effective methods to estimate defectiveness in\nsource code. However, the existing methods are dependent on experts or\ninefficient. In this paper, we improve the performance (in terms of speed and\nmemory usage) of our existing code review assisting tool--CRUSO. The central\nidea of the approach is to estimate the defectiveness for an input source code\nby using the defectiveness score of similar code fragments present in various\nStackOverflow (SO) posts.\n The significant contributions of our paper are i) SOpostsDB: a dataset\ncontaining the PVA vectors and the SO posts information, ii) CRUSO-P: a code\nreview assisting system based on PVA models trained on \SOpostsDB. For a\ngiven input source code, CRUSO-P labels it as Likely to be defective, Unlikely\nto be defective, Unpredictable. To develop CRUSO-P, we processed >3 million SO\nposts and 188200+ GitHub source files. CRUSO-P is designed to work with source\ncode written in the popular programming languages C, C#, Java, JavaScript, and\nPython.\n CRUSO-P outperforms CRUSO with an improvement of 97.82% in response time and\na storage reduction of 99.15%. CRUSO-P achieves the highest mean accuracy score\nof 99.6% when tested with the C programming language, thus achieving an\nimprovement of 5.6% over the existing method.\n