Applying Multiple Classifier Combination by Clustering and Selection to SoftMan´s Perception System

SoftMan, the virtual robot in network environment, is a kind of software artificial life living in computer networks. It has a humanoid structure, and simulating human, its perception system should be able to recognize perceptive objects. Enlightened by human\´s perception system and the "disassemble-integration" method in analyzing large systems, a cooperative classification model in SoftMan\´s perception system is proposed. In the model, different humanoid senses are simulated by multiple classifier systems (MCS). An essential issue in the research of MCS is multiple classifier combination. Accordingly, a method of multiple classifier combination is proposed and validated in this paper.