A Systematic Review of Unsupervised Learning Techniques for Software Defect Prediction
share
Background: Unsupervised machine learners have been increasingly applied to software defect prediction. It is an approach that may be valuable for software practitioners because it reduces the need for labeled training data. Objective: Investigate the use and performance of unsupervised learning techniques in software defect prediction. Method: We conducted a systematic literature review that identified 48 studies containing 2348 individual experimental results, which satisfied our inclusion criteria published between January 2000 and March 2018. In order to compare prediction performance across these studies in a consistent way, we (re-)computed the confusion matrices and employed Matthew's correlation coefficient (MCC) as our main performance measure. Results: Our meta-analysis shows that unsupervised models are comparable with supervised models for both within-project and cross-project prediction. Among 21 unsupervised models, Fuzzy CMeans (FCM) and Fuzzy SOMs (FSOMs) perform best. In addition, where we were able to check, we found that almost 11 published results (contained in 16 papers) were internally inconsistent and a further 30 Conclusion: Although many factors impact the performance of a classifier, e.g., dataset characteristics, broadly speaking, unsupervised classifiers do not seem to perform worse than the supervised classifiers in our review. However, we note a worrying prevalence of (i) demonstrably erroneous experimental results, (ii) undemanding benchmarks and (iii) incomplete reporting. We particularly encourage researchers to be comprehensive in their reporting.
READ FULL TEXT
