• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2023 Fiscal Year Final Research Report

Online Optimization of Defect Prediction Models Towards High Quality Software

Research Project

  • PDF
Project/Area Number 21K11840
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 60050:Software-related
Research InstitutionKindai University

Principal Investigator

Tsunoda Masateru  近畿大学, 情報学部, 准教授 (60457140)

Project Period (FY) 2021-04-01 – 2024-03-31
Keywords多腕バンディット問題 / 機械学習 / ソフトウェア開発プロジェクト / 性能評価
Outline of Final Research Achievements

To enhance the quality of software, we have focused on the online optimization of defect prediction models. The achievements of this project are classified into three types: (1) accuracy improvement of defect prediction models, (2) application of online optimization besides defect prediction, and (3) improvement of online optimization. Specifically, (1) involves online optimization to select better methods such as variable reduction methods and ensemble learning methods. (2) applies online optimization to activities such as code clone detection, code generation, and software review. (3) clarifies the problem of online learning and its improvement, and proposes a new approach to software testing. The achievements of our project are expected to bring high-quality software.

Free Research Field

ソフトウェア工学

Academic Significance and Societal Importance of the Research Achievements

ソフトウェア欠陥予測において汎用的なモデルや手法は存在しない.このため,平均的に性能の高い予測方法などを事前に評価する研究が広く行われてきた.このような従来のアプローチでは(1) 事前に様々な手法を評価する必要があり,評価のための時間とコストが掛かる,(2) 平均的に性能の高い手法が,適用対象のプロジェクトで性能が高いとは限らない,という問題点があった.本研究のオンライン最適化により(1)の事前評価が必須ではなくなり,新しい手法を積極的に利用可能とした,(2) 適用対象のプロジェクトにおいて性能が低下するリスクを避けることができ,予測方法などを実プロジェクトに積極的に導入することを可能にした.

URL: 

Published: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi