| | Study | Architecture | Combining approach |
Ensemble learning phase and ensemble level |
Combining method employed |
Metric |
Dataset | Diversity | Base classifier | | Generation | Selection | Integration |
| | Giacinto and Roli [45] | Parallel | Ensemble | Feature level | — | Fusion | Majority voting, average rule, belief function | Error rate, FPR, cost | KDD 99 | Implicit | NN | | Sabhnani and Serpen [23] | — | Hybrid | — | Classifier level | — | Multi-classifiers method | DR, FPR | KDD 99 | — | NN, KM, GC | | Chebrolu et al. [28] | Parallel | Ensemble | — | Classifier level | Selection | Weighting method | CA | KDD 99 | Implicit | BN, CART | | Abraham et al. [43] | Parallel | Ensemble | Feature level | Classifier level | Selection | Weighting method | CA | KDD 99 | Implicit | DT, SVM | | Kruegel et al. [109] | Parallel | Ensemble | Feature and data level | — | Fusion | Score-and probability-based method | FPR | Real world dataset | Implicit | BN | | Perdisci et al. [88] | — | Ensemble | — | — | Fusion | Clustering | — | Real world dataset | — | — | | Hwang et al. [42] | Cascading | Hybrid | — | — | — | Consecutive combination | DR, FPR | KDD 99 | — | SVM | | Chen et al. [41] | Hierarchical | Hybrid | Feature level | — | — | Multi-classifiers method | DR, FNR, FPR | KDD 99 | — | FNT | | Khan et al. [40] | Cascading | Hybrid | — | — | — | Clustering + classification | CA, training time, FP, FN | KDD 99 | — | SVM, clustering | | Toosi and kahani [39] | Parallel | Ensemble | — | Classifier level | Fusion | Fuzzy theory method | CA, DR, FPR, CPE | KDD 99 | Implicit | NN, fuzzy logic | | Yan and Hao [111] | Parallel | Ensemble | Feature level | — | Selection | — | DR, FPR | KDD 99 | Implicit | NN | | Xiang et al. [36] | Cascading | Hybrid | Data level | Classifier level | — | Clustering + classification | TP, FP | KDD 99 | — | DT, BC
| | Cretu et al. [113] | Parallel | Ensemble | Data level | — | Fusion | Voting method | FP, TP | Real world data | — | Anagram, Payl
| | Hu et al. [112] | Parallel | Ensemble | Feature level | — | — | Mixture of expert systems | DR, FAR, computation time | KDD 99 | Implicit | DS
| | Corona et al. [110] | Parallel | Ensemble | Feature and data level | — | Fusion | Threshold probability method | FPR, DR | Real world dataset | Implicit | HMM | | Zainal et al. [35] | Parallel | Ensemble | Feature level | Classifier level | Fusion | Weighted voting method | CA, TP, FP | KDD 99 | Implicit | LGP, ANFIS, RF
| | Menahem et al. [106] | Parallel | Ensemble | Data level | Classifier level | Fusion | Meta learning | CA, area under the ROC curve, training time | Real-time network traffic | Implicit | DT, NB, K-NN, VFI, OneR
| | Wang et al. [32] | Parallel | Ensemble | Data level | — | Fusion | Meta learning | Precision, recall, F-measure | KDD 99 | Implicit | NN, fuzzy logic, clustering | | Khreich et al. [2] | Parallel | Ensemble | — | — | Fusion | Iterative Boolean combination method | ROC space | UNM dataset, real world dataset | Implicit | HMM | | Govindarajan and Chandrasekaran [30] | Parallel | Ensemble | Data level | — | Fusion | Weighted method | CA | Immune system dataset from University of New Mexico | Implicit | MLP, RBF | | Muda et al. [120] | Cascading | Hybrid | Data level | — | — | Clustering + classification | CA, DR, FPR | KDD 99 | — | KM, NB |
|
|
