Adaptive Prefix Filtering for Accurate Code Clone Detection in Conjunction with Meta-learning

• Published in: SN computer science Vol. 5; no. 6; p. 789
• Main Authors: Ralhan, Chavi, Malik, Navneet, Agrawal, Prateek, Gupta, Charu, jatana, Nishtha, Jatain, Divya, Sharma, Geetanjali
• Format: Journal Article
• Language: English
• Published: Singapore Springer Nature Singapore 12.08.2024; Springer Nature B.V
• Subjects: Algorithms; Artificial intelligence; Cloning; Computer Imaging; Computer Science; Computer Systems Organization and Communication Networks; Data points; Data Structures and Information Theory; Datasets; Deep learning; Effectiveness; Efficiency; Emerging Applications of Cyber-Physical System; Filtration; Information Systems and Communication Service; Innovations; Machine learning; Neural networks; Optimization techniques; Original Research; Pattern Recognition and Graphics; Programming languages; Python; Research projects; Software development; Software Engineering/Programming and Operating Systems; Syntax; Vision; Java; Abstract prefix filtering; Meta-learning; Programming languages; Code clone detection; Python
• ISSN: 2661-8907; 2662-995X; 2661-8907
• DOI: 10.1007/s42979-024-03140-5

This research project has significantly contributed to the advancement of code duplication detection by introducing a new testing method and constructing a highly accurate meta-classifier. The evaluation of various architectures and utilization of a diverse dataset enabled the development of a novel and versatile solution for detecting duplicate code in both Java and Python programming languages. The proposed algorithm relies on a set of distinctive qualities represented by 17 code metrics. It efficiently stores the code structure in Abstract Prefix filtering (APF) to identify duplicate code segments. A comprehensive series of experiments is conducted to evaluate the effectiveness of different architectures for meta-classifiers. The goal was to identify an optimal 2-layer stacking meta-classifier to yield precise and reliable results. This process successfully constructed a novel classifier, demonstrating exceptional accuracy in detecting duplicate code. The algorithm was trained on a dataset comprising 19,988 data points, encompassing code metrics from both Java and Python programming languages. This diverse dataset enabled the model to learn and generalize across multiple language paradigms, enhancing its versatility and effectiveness in code clone detection. The results show that the proposed model outperformed the state-of-the-art models, which proves that it is the appropriate choice for constructing a meta-classifier for cloned code detection.