… More specifically, we compared the top performing classifiers we compare … benchmarking
study shows that software defect prediction should be assessed using extensive evaluation …

… published papers evaluate the cost efficiency of predictions, ie… RQ1, we compare the mean
performance of each approach … this benchmark, we consider 67 products, which is large for …

… to assess the performance GIS in comparison with benchmark CPDP (… To assess the impact
of feature sets, we use two sets of … , benchmarks and results, requires more comprehensive …

… problem in comparative studies of prediction models. To assess the risk that defect prediction
… Hence, very few correct (or incorrect) predictions can have a very large impact on cross-…

… model having higher accuracy and suitable for a large-scale … essential to find out ways and
means for defect prediction. By an … that, we evaluated and compared the obtained prediction …

… testing effort in this paper) of applying fault-prone prediction models during evaluation and
… , “Evaluating defect prediction approaches: a benchmark and an extensive comparison,” …

… differences of 40 clustering-based unsupervised models (as well as 6 supervised models for
comparison) on a public benchmark … We evaluated these methods with one traditional and 2 …

… to compare the … approach the prediction task using very different techniques. From each
category of system in our analysis, we observe a considerable number of defects predicted by a …

… this paper is to provide an extensive comparison of well-known … of cross-project defect
prediction significantly, and the … , “Benchmarking classification models for software defect …

… at review time makes a large difference in helping developers as opposed to producing the
… performance evaluation (eg, we see a very healthy and progressive trend where software …