Due to its large dimensionality, it is implemented using machine learning tool, ensemble classifier [16] which consists of many base learners such as FLD [29].
The classification performance of our proposed algorithms is restricted by many factors, including the size of feature space, the size of feature subspace, and the number of base classifiers; the size of feature subspace is the most significant factor.
Since many gene expression based classifiers are developed retrospectively, there is often little control of the sample size.
Successful classifier combination requires diversity of the base classifiers.
Cascade: combine base classifiers according to their final weights and construct strong classifier.
Finally, the strong classifier is constructed similar with AdaBoost according to base classifiers' weight.
Multiple base classifiers are built in AntMinermbc, and each base classifier is expected to remedy the weakness of other base classifiers, which can improve the predictive accuracy by exploiting the useful information from various base classifiers.
Three schemes are used to fuse these base classifiers.
Type-1 architecture is composed of homogeneous base classifiers and Type-2 architecture is constructed using heterogeneous base classifiers.
