Investigating data-flow coverage of classes using evolutionary algorithms

It is not unusual for a software development organization to expend 40 percent of total project effort on testing, which can be a very laborious and time-consuming process. Thus, there is a big necessity for test automation. This paper describes an approach to automatically generate test-data for the evolutionary testing of object-oriented software exploiting data-flow coverage. An implementation that is a proof-of-concept for the proposed technique is also presented, and the experimental results obtained by employing this tool for the unit testing of six real-world Java classes are discussed and compared with existing studies in the field, when possible. Additionally, the analysis of these results leads to useful findings that suggest that evolutionary testing can be utilized successfully under the object-oriented paradigm, dataflow coverage provides more thorough testing but with a higher computational cost, and that the proposed technique has a few flaws that need to be addressed i...

• Type: Conference
• Year: 2007
• Where: GECCO
• Authors: Konstantinos Liaskos, Marc Roper, Murray Wood
• Keywords: Data-flow Coverage, Evolutionary Testing, GECCO 2007, Keywords Object-oriented Testing