Design and Analysis of Novel Kernel Measure for Software Fault Localization

Abstract: The problem of software fault localization may be viewed as an approach for finding hidden faults or bugs in the existing program codes which are syntactically correct and give fault free output for some input instances but fail for all other input instances. Some of the reasons include logical errors, wrong interpretation of specification, coding errors. Finding such faults is not possible sometimes with the help of compilers. This is where the necessity and significance of software fault localization stems out. The main contribution for this work is to first introduce the block hit-miss function which relates block vectors of execution sequences of software code over sample runs performed and the decision vector which denotes fault or error free output. The similarity measure is applied to the block vector and decision vectors as input and the pair with maximum similarity is considered as faulty block.