Applied Deep Code Search for Algorithm Implementation using Pseudocode

Thumbnail Image
Date
2022-05
Authors
Valluri, Sai Charishma
Major Professor
Le, Wei
Advisor
Committee Member
Li, Qi
Journal Title
Journal ISSN
Volume Title
Publisher
Authors
Research Projects
Organizational Units
Journal Issue
Is Version Of
Versions
Series
Department
Abstract
Code retrieval tools and techniques play a key role in facilitating the software developers to retrieve code snippets from open-source projects given a natural language query. With natural language description as an input, the code search tool searches for the most relevant code snippets amongst the code. This creative component focuses on training the state-of-the-art code search tool, Deep Code Search, with the dataset of algorithms as a source code and studying the resultant code snippets when given the pseudo-code as an input query. We trained the Deep Code Search for three different programming language settings, Java, C, and a combination of Java and C. Our results show that Deep Code Search can identify the algorithm implementation in different languages for the given pseudo-code.
Comments
Description
Keywords
Citation
DOI
Source
Copyright
2022