Software development support using source code corpus

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K00108
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Software
• Research Institution: Nihon University
• Principal Investigator: YAMAMOTO Tetsuo 日本大学, 工学部, 准教授 (40388129)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: コード補完 / コード推薦 / RNN

Outline of Final Research Achievements

Developers reuse existing source code or use libraries to develop effectively. In this study, we focus on the order of method invocation statements in existing source code and propose to suggest method invocation statements. This study proposed an approach to suggest method invocation statements using recurrent neural network. I have implemented the approach and conducted experiments to measure an accuracy with 10 open source software projects. I have investigated various parameters of recurrent neural network. This evaluation has shown that our approach is 38% accuracy in API code suggestion, it can correctly suggest the API with top 1 candidate.

Free Research Field

ソフトウェア工学