Cluster-based fuzzy regression trees for software cost prediction

The current paper <span lang="EN-US">proposes a novel type of decision tree, which is never used for software development cost prediction (SDCP) purposes, the cluster-based fuzzy regression tree (CFRT). This model uses k-means (FKM), deals with data uncertainty and imprecision. expansion based on variability measure by choosing node highest value granulation diversity. outlined an experimental study comparing CFRT four SDCP methods, notably linear regression, multi-layer perceptron, K-nearest-neighbors, classification trees (CART), employing eight datasets leave-one-out cross-validation (LOOCV). results show that among best, ranked first in 3 according to accuracy measures. Also, Pred(25%) values, proposed outperformed all twelve compared techniques datasets: Albrecht, constructive (COCOMO), Desharnais, International Software Benchmarking Standards Group (ISBSG) using LOOCV 30-fold technique.</span>