Table 1: Comparison of related works on the use of ensemble models for software effort estimation.

StudyEnsemble typeBase learner(s)Combination rule(s)Number of datasets

Braga et al. [22]Homogeneous
(bagging)
Linear regressionLinear (averaging) 1
Homogeneous
(bagging)
MLPLinear (averaging)
Homogeneous
(bagging)
M5P regression treesLinear (averaging)
Homogeneous
(bagging)
M5P model treesLinear (averaging)
Homogeneous
(bagging)
SVRLinear (averaging)

Kultur et al. [23, 24]Homogeneous
(bootstrapping)
MLPNonlinear (average of largest cluster obtained using adaptive resonance theory (ART) algorithm)5

Minku and Yao [25, 26]Homogeneous
(bagging, random, negative correlation learning)
MLPLinear (averaging) 18
Homogeneous
(bagging)
RBFLinear (averaging)
Homogeneous
(bagging)
Regression TreesLinear (averaging)

Kocaguneli et al. [27]HeterogeneousGaussian process, MLP, RBF, SMOReg, SVMReg, IBk, LWL, additive regression, bagging with decision tree, RandomSubSpace, DecisionStump, M5P, ConjunctiveRule, DecisionTableLinear (averaging)3

Kocaguneli et al. [28]HeterogeneousABE0-1NN, ABE0-5NN, SWReg, CART (yes), CART (no), NNet, LReg, PCR, PLSRLinear (mean, median, inverse-ranked weighted mean (IRWM))20

Elish [29]HeterogeneousMLP, RBF, RT, KNN, SVRLinear (median)5

Homogeneous (bagging)MLPLinear (averaging, weighted averaging) and nonlinear (MLP
SVR, FIS-FCM, FIS-SC, ANFIS-FCM, ANFIS-SC)
This paperHomogeneous
(bagging)
SVR 5
Homogeneous
(bagging)
ANFIS
HeterogeneousMLP, SVR, ANFIS