Conjoint data is data in which the classes abut but do not overlap. It is difficult to determine the boundary between the classes as there are no inherent clusters in conjoint data and as a result traditional classification methods, such as counter propagation networks, may under perform. This paper describes a modified counter propagation network that is able to refine the boundary definition and so perform better when classifying conjoint data. The efficiency with which it uses the network resources suggests that it is worthy of consideration for classifying all kinds of data.
展开▼