Bootstrapping and aggregation of classifiers
W = BAGGINGC (A,CLASSF,N,ACLASSF,T)
Computation of a stabilised version of a classifier by bootstrapping and aggregation ('bagging'). In total N bootstrapped versions of the dataset A are generated and used for training of the untrained classifier CLASSF. Aggregation is done using the combining classifier specified in ACLASSF. If ACLASSF is a trainable classifier it is trained by the tuning dataset T, if given; else A is used for training. The default aggregating classifier ACLASSF is MAXC. Default base classifier CLASSF is NMC.
In case the aggregating classifier ACLASSF is WVOTEC the weights for the voting are derived from the apparent errors based on the bootstrapped versions of the training set A.
In multi-class problems another way of combining might be of interest
L.Beiman, Bagging Predictors, Machine Learning, 1996.