Software fault prediction means identification of the faultprone parts in the software. This enables to focus testing activities on those software modules that are predicted as fault-prone. As can be seen in literature, many soft computing techniques are employed to make more accurate predictions previously. However, software fault prediction has not become routine activity in the software development process, because most of employed methods require historical data to train the model. In fact, collection of the historical data is not a simple job and also collected data represents the project which was observed. It may not be reusable for different projects. To overcome these problems, use of Mamdani type fuzzy inference system to predict software fault prone modules is suggested in this study. Another reason is to eliminate the disadvantages sourced from the small size of data. In this study, object oriented metrics are preferred because of widespread use of object oriented technologies. Experimental results show that fuzzy inference systems are successful and can be competitive with methods previously employed in the literature.
KEY WORDSSoftware fault prediction, object-oriented metrics, fuzzy inference systems.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.