Tool for Improved Software Bug Localization
Purdue University researchers have developed an algorithm that can significantly narrow down the portion of the software that must be examined to create a bug fix. The method mines the version histories of the software artifacts created incrementally during software development. It employs a novel Information Retrieval framework that leverages the relative importance of each incremental change of the software to a given bug using the change's associated time stamp in the repository logs. This method has been shown to greatly improve bug localization accuracy.
Significantly increases accuracy of bug localizationAchieves an accurate estimation of the relative importance of each file to a particular bug in a software repository
Computer TechnologySoftware
Avinash KakPurdue Electrical and Computer Engineering
United States
None
美國
