Skip to main navigation Skip to search Skip to main content

Software complexity measurement based on complex network

  • Beihang University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Usually the complexity metric of software focuses on the complexity of code level, function level or structure level separately. It lacks of measurement for the comprehensive complexity of software system. This paper proposes a complexity metric model of three-level cascade network that based on complex network theory. In this metric model, the complexity of code level, function level and structure level are measured and the cascaded relationship between the three levels are analyzed. At last, the three-level cascade network model is built and the comprehensive complexity of software system is measured though the three-level cascade network model. The experiment result shows that the comprehensive complexity of the software system is correlated positively to the number of software defects.

Original languageEnglish
Title of host publicationICSESS 2017 - Proceedings of 2017 IEEE 8th International Conference on Software Engineering and Service Science
EditorsLi Wenzheng, M. Surendra Prasad Babu, Lei Xiaohui
PublisherIEEE Computer Society
Pages262-265
Number of pages4
ISBN (Electronic)9781538645703
DOIs
StatePublished - 2 Jul 2017
Event8th IEEE International Conference on Software Engineering and Service Science, ICSESS 2017 - Beijing, China
Duration: 24 Nov 201726 Nov 2017

Publication series

NameProceedings of the IEEE International Conference on Software Engineering and Service Sciences, ICSESS
Volume2017-November
ISSN (Print)2327-0586
ISSN (Electronic)2327-0594

Conference

Conference8th IEEE International Conference on Software Engineering and Service Science, ICSESS 2017
Country/TerritoryChina
CityBeijing
Period24/11/1726/11/17

Keywords

  • cascade network
  • complex network
  • defect prediction
  • software complexity
  • software metric

Fingerprint

Dive into the research topics of 'Software complexity measurement based on complex network'. Together they form a unique fingerprint.

Cite this