Research indexConference paper
Code Clone Detection Method Using Vector Representation Based on Information Retrieval Technology and Deep Learning Model
Adhikari, Kumar, Deshwal
28 Nov 2025
NobleIDark:/48914/w1/20261105/2C6CE4B8DOI10.1109/emergin67762.2025.11450760
Code Clone Detection Method Using Vector Representation Based on Information Retrieval Technology and Deep Learning Model
Adhikari, Kumar, Deshwal · 2025 · Subscription
NobleIDark:/48914/w1/20261105/2C6CE4B8DOI10.1109/emergin67762.2025.11450760
The full text is at the publisher
You can read the abstract here, or open it at the publisher.
Abstract
Code clone detection is a fundamental technique in software engineering used to identify duplicate code segments. While many methods exist for detecting syntactic similarities, few effectively capture semantic similarity. DeepSim, a deep learning- based approach, achieves high semantic accuracy but suffers from significant computational latency due to its prolonged model training time. Alternatively, TF-IDF-based methods offer faster detection but struggle with low syntactic similarity clones. To address these issues, we propose a hybrid method that combines information retrieval based vector representation with deep learning to achieve faster and more accurate code similarity detection.