A quantitative and qualitative investigation of performance-related commits in android apps

Teerath Das, Massimiliano Di Penta, Ivano Malavolta

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

17 Downloads (Pure)

Abstract

Performance is nowadays becoming a crucial issue for mobile apps, as they are often implementing computational-intensive features, are being used for mission-critical tasks, and, last but not least, a pleasant user experience often is a key factor to determine the success of an app. This paper reports a study aimed at preliminarily investigating to what extent developers take care of performance issues in their commits, and explicitly document that. The study has been conducted on commits of 2,443 open source Android apps, of which 180 turned out to contain a total of 457 documented performance problems. We classified performance-related commits using a card sorting approach, and found that the most predominant kinds of performance-related changes include GUI-related changes, fixing code smells, network-related code, and memory management.

Original languageEnglish
Title of host publication2016 IEEE International Conference on Software Maintenance and Evolution (ICSME)
Subtitle of host publication[Proceedings]
PublisherInstitute of Electrical and Electronics Engineers, Inc.
Pages443-447
Number of pages5
ISBN (Electronic)9781509038060
ISBN (Print)9781509038077
DOIs
Publication statusPublished - 2017
Event32nd IEEE International Conference on Software Maintenance and Evolution, ICSME 2016 - Raleigh, United States
Duration: 2 Oct 201610 Oct 2016

Conference

Conference32nd IEEE International Conference on Software Maintenance and Evolution, ICSME 2016
Country/TerritoryUnited States
CityRaleigh
Period2/10/1610/10/16

Keywords

  • Android
  • App store mining
  • Mobile performance issues

Fingerprint

Dive into the research topics of 'A quantitative and qualitative investigation of performance-related commits in android apps'. Together they form a unique fingerprint.

Cite this