A novel approach to collaborative testing in a crowdsourcing environment

Software testing processes are generally labor-intensive and often involve substantial collaboration among testers, developers, and even users. However, considerable human resource capacity exists on the Internet in social networks, expert communities, or internet forums—referred to as crowds. Effec...

Full description

Saved in:
Bibliographic Details
Published in:The Journal of systems and software 2013-08, Vol.86 (8), p.2143-2153
Main Authors: Tung, Yuan-Hsin, Tseng, Shian-Shyong
Format: Article
Language:English
Subjects:
Assignment problem
Cloud computing
Collaboration
Collaborative testing
Crowdsourcing
Heuristic
Integer linear programming
Integer programming
Linear programming
Software engineering
Software testing
Studies
Test systems
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Software testing processes are generally labor-intensive and often involve substantial collaboration among testers, developers, and even users. However, considerable human resource capacity exists on the Internet in social networks, expert communities, or internet forums—referred to as crowds. Effectively using crowd resources to support collaborative testing is an interesting and challenging topic. This paper defines the collaborative testing problem in a crowd environment as an NP-Complete job assignment problem and formulates it as an integer linear programming (ILP) problem. Although package tools can be used to obtain the optimal solution to an ILP problem, computational complexity makes these tools unsuitable for solving large-scale problems. This study uses a greedy approach with four heuristic strategies to solve the problem. This is called the crowdsourcing-based collaborative testing approach. This approach includes two phases, training phase and testing phase. The training phase transforms the original problem into an ILP problem. The testing phase solves the ILP using heuristic strategies. A prototype system, called the Collaborative Testing System (COTS), is also implemented. The experiment results show that the proposed heuristic algorithms produce good quality approximate solutions in an acceptable timeframe.
ISSN:0164-1212
1873-1228
DOI:10.1016/j.jss.2013.03.079