Current browse context:
cs.LG
Change to browse by:
References & Citations
Computer Science > Machine Learning
Title: MISIM: A Neural Code Semantics Similarity System Using the Context-Aware Semantics Structure
(Submitted on 5 Jun 2020 (v1), last revised 2 Jun 2021 (this version, v6))
Abstract: Code semantics similarity can be used for many tasks such as code recommendation, automated software defect correction, and clone detection. Yet, the accuracy of such systems has not yet reached a level of general purpose reliability. To help address this, we present Machine Inferred Code Similarity (MISIM), a neural code semantics similarity system consisting of two core components: (i)MISIM uses a novel context-aware semantics structure, which was purpose-built to lift semantics from code syntax; (ii)MISIM uses an extensible neural code similarity scoring algorithm, which can be used for various neural network architectures with learned parameters. We compare MISIM to four state-of-the-art systems, including two additional hand-customized models, over 328K programs consisting of over 18 million lines of code. Our experiments show that MISIM has 8.08% better accuracy (using MAP@R) compared to the next best performing system.
Submission history
From: Shengtian Zhou [view email][v1] Fri, 5 Jun 2020 20:55:23 GMT (9161kb,D)
[v2] Mon, 15 Jun 2020 19:56:43 GMT (9161kb,D)
[v3] Tue, 6 Oct 2020 19:39:45 GMT (9509kb,D)
[v4] Fri, 9 Oct 2020 00:25:35 GMT (9509kb,D)
[v5] Mon, 22 Feb 2021 23:46:57 GMT (11982kb,D)
[v6] Wed, 2 Jun 2021 20:34:31 GMT (12258kb,D)
Link back to: arXiv, form interface, contact.