A Challenge toward Automated Source Code Evolution

2014 Fiscal Year Final Research Report

• Project/Area Number: 24650011
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Software
• Research Institution: Osaka University
• Principal Investigator: HIGO Yoshiki 大阪大学, 情報科学研究科, 准教授 (70452414)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: リポジトリマイニング / ソースコード解析 / 機械学習

Outline of Final Research Achievements

In this research, we have developed a model to predict the next change on a given Java method. The model is built based on the past code changes on the target software system. The model adopts a vector data, which consists of the number of every elements in the source code. For example, binomial expressions and return statements are such elements. We conducted experiments with the model and confirmed that the model was able to predict the next changes correctly with 75--85% accuracy.

Free Research Field

ソフトウェア工学