Title: Bug triaging based on ant systems

Authors: V. Akila; G. Zayaraz; V. Govindasamy

Addresses: Department of Computer Science and Engineering, Pondicherry Engineering College, Pondicherry, India ' Department of Computer Science and Engineering, Pondicherry Engineering College, Pondicherry, India ' Department of Information Technology, Pondicherry Engineering College, Pondicherry, India

Abstract: The open source software development model is a highly complex activity among voluntary software developers. This model of development has led to the availability of software repositories in the public domain. Bug repositories are part of software repositories. They maintain and store bug reports that arise after software has been deployed. Bug triaging, typically is a recommendation problem in a volatile environment. Bug triaging systems use the past history of bug re-assignments and information present in the bug report summary to make new recommendations. Any recommendation system needs positive and negative feedback on the recommendation made so as to fine tune its performance and to remain relevant. Ant colony optimisation is a stochastic technique that has pheromone trail deposit as positive feedback and pheromones evaporation as negative feedback. This paper presents a framework for automated bug triaging system based on ant colony optimisation. The performance of the proposed framework is validated with the bug reports of Eclipse Project. The parameters including: 1) path length; 2) precision; 3) recall; 4) path similarity are considered. The performance evaluation shows that the proposed framework is better than the existing weighted breadth first search (WBFS) system.

Keywords: automated bug triaging; ant colony optimisation; ACO; open source software; OSS development; software development; bug repositories; software repositories; software bugs; path length; precision; recall; path similarity.

DOI: 10.1504/IJBIC.2015.071078

International Journal of Bio-Inspired Computation, 2015 Vol.7 No.4, pp.263 - 268

Received: 01 Nov 2014
Accepted: 22 Dec 2014

Published online: 11 Aug 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article