Search: onr:"swepub:oai:DiVA.org:kth-312782" >
Multimodal Represen...
Multimodal Representation for Neural Code Search
-
- Gu, Jian (author)
- Univ Zurich, Zurich, Switzerland.
-
- Chen, Zimin (author)
- KTH,Teoretisk datalogi, TCS
-
- Monperrus, Martin (author)
- KTH,Teoretisk datalogi, TCS
-
Univ Zurich, Zurich, Switzerland Teoretisk datalogi, TCS (creator_code:org_t)
- Institute of Electrical and Electronics Engineers (IEEE), 2021
- 2021
- English.
-
In: 2021 IEEE international conference on software maintenance and evolution (ICSME 2021). - : Institute of Electrical and Electronics Engineers (IEEE). ; , s. 483-494
- Related links:
-
https://urn.kb.se/re...
-
show more...
-
https://doi.org/10.1...
-
show less...
Abstract
Subject headings
Close
- Semantic code search is about finding semantically relevant code snippets for a given natural language query. In the state-of-the-art approaches, the semantic similarity between code and query is quantified as the distance of their representation in the shared vector space. In this paper, to improve the vector space, we introduce tree-serialization methods on a simplified form of AST and build the multimodal representation for the code data. We conduct extensive experiments using a single corpus that is large-scale and multi-language: CodeSearchNet. Our results show that both our tree-serialized representations and multimodal learning model improve the performance of code search. Last, we define intuitive quantification metrics oriented to the completeness of semantic and syntactic information of the code data, to help understand the experimental findings.
Subject headings
- SAMHÄLLSVETENSKAP -- Ekonomi och näringsliv -- Företagsekonomi (hsv//swe)
- SOCIAL SCIENCES -- Economics and Business -- Business Administration (hsv//eng)
- SAMHÄLLSVETENSKAP -- Social och ekonomisk geografi -- Kulturgeografi (hsv//swe)
- SOCIAL SCIENCES -- Social and Economic Geography -- Human Geography (hsv//eng)
- TEKNIK OCH TEKNOLOGIER -- Elektroteknik och elektronik -- Kommunikationssystem (hsv//swe)
- ENGINEERING AND TECHNOLOGY -- Electrical Engineering, Electronic Engineering, Information Engineering -- Communication Systems (hsv//eng)
Keyword
- multimodal learning
- program representation
- information completeness
- tree serialization
- code search
Publication and Content Type
- ref (subject category)
- kon (subject category)
To the university's database