DESIGN AND IMPLEMENTATION OF A HYBRID MULTI CONCEPTS ACQUISITION SYSTEM

This paper describes an incremental hybrid multi concept acquisition system HMCAS.It puts forward a concept learning algorithm based on the probability theory,which is a hybrid of symbolic method and neural network.HMCAS conjoins the symbolic and neural network approaches tightly,and it has an incremental learning mechanism.Users can select BP or FTART approach to accomplish neural network learning.Its inference mechanism has the ability of hybrid decision tree inference.