The accuracy of information retrieval based bug localization techniques

Bug localization involves using information about a bug to locate affected code sections. Several automated bug localization techniques based on information retrieval (IR) models have been constructed in recent years. Latent semantic indexing and latent Dirichlet allocation are two such techniques.;The "gold standard" for measuring an IR technique's accuracy considers the technique's ability to locate a "first relevant method." However, the question remains -- does finding this single method enable location of a complete set of bug related methods? Previous arguments assume this to be true; however, few analyses of this assumption have been performed. Our first task is to test this assumption.;To further measure IR accuracy, we analyze the relevancy of the IR model's "first method returned." We use various structural analysis techniques to extend relevant methods and determine accuracy and reliability of these assumptions. We also compare the accuracy of both IR techniques in this context.