Development of fault-prone prediction model using web cloud system

2014 Fiscal Year Final Research Report

• Project/Area Number: 24500038
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Software
• Research Institution: Kyoto Institute of Technology
• Principal Investigator: MIZUNO Osamu 京都工芸繊維大学, 工芸科学研究科, 准教授 (60314407)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: ソフトウェア / ソースコード / リポジトリマイニング / 識別子 / トピック分析

Outline of Final Research Achievements

In this research, we proposed a system to predict fault-prone modules in software source code. To do so, we focused on the vocabulary of the identifiers in source code. First, we developed a system to find topics in the faulty source code. Second, we developed a system to restore abbreviated identifiers to the original identifiers using a deep learning technique. By these models, we found the ways to utilise identifiers in software for fault-prone module prediction.

Free Research Field

ソフトウェア工学