Skip to main navigation Skip to search Skip to main content

EPR: a Neural Network for Automatic Feature Learning from Code for Defect Prediction

  • Dingbang Fang
  • , Shaoying Liu*
  • , Ai Liu
  • *Corresponding author for this work
  • Hiroshima University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Software defect prediction plays a significant role in the software development cycle but suffers from many difficulties. In this paper we propose a novel deep learning model (including algorithms) called Extractor, Parser, and Reviewer (EPR) for defect prediction in software. Two different networks, recurrent neural networks (RNNs) and one-dimensional convolutional networks(ODCNs), are employed by the EPR for different purposes. RNN is utilized to extract contextual features to represent semantic dependencies between code tokens and ODCN acts as a parser to establish dependencies between semantic features. Meanwhile, the attention mechanism of the two networks is used as a reviewer to assign different weights from location information to the importance of the features, respectively. Our proposed model is validated by the PROMISE repository, and the results show that the proposed model in this paper significantly outperforms several existing algorithms.

Original languageEnglish
Title of host publicationProceedings - 2021 21st International Conference on Software Quality, Reliability and Security, QRS 2021
PublisherInstitute of Electrical and Electronics Engineers
Pages482-492
Number of pages11
ISBN (Electronic)9781665458139
DOIs
StatePublished - 2021
Externally publishedYes
Event21st International Conference on Software Quality, Reliability and Security, QRS 2021 - Hainan, China
Duration: 6 Dec 2021 → 10 Dec 2021

Publication series

NameIEEE International Conference on Software Quality, Reliability and Security, QRS
Volume2021-December
ISSN (Print)2693-9177

Conference

Conference21st International Conference on Software Quality, Reliability and Security, QRS 2021
Country/TerritoryChina
CityHainan
Period6/12/21 → 10/12/21

Keywords

  • attention mechanism
  • defect prediction
  • neural network

Fingerprint

Dive into the research topics of 'EPR: a Neural Network for Automatic Feature Learning from Code for Defect Prediction'. Together they form a unique fingerprint.

Cite this