Nonlinear committee pattern classification

Methods which combine outputs of multiple pattern classifiers to enhance the overall classification rate are studied. Specific attention is given to combination rules which are independent of the input feature vectors. Potential performance enhancement and limits of this so called stack generalization method are discussed. In particular, a phenomenon called “alias” is introduced which gives an upper bound of the performance which can be achieved using stack generation for a given set of member classifiers. Experimentation using several machine learning databases are reported