Incremental learning of fuzzy rule-based classifiers for large data sets

Incremental construction of fuzzy rule-based classifiers is studied in this paper. It is assumed that not all training patterns are given a priori for training classifiers, but are gradually made available over time. It is also assumed the previously available training patterns can not be used in the following time steps. Thus fuzzy rule-based classifiers should be constructed by updating already constructed classifiers using the available training patterns at each time step. Two methods are proposed for the incremental construction of fuzzy rule-based classifiers. The first method updates the fuzzy if-then rules by considering individual training patterns separately while in the second method all available training patterns are used together in the update procedure of the fuzzy if-then rules. A series of computational experiments are conducted in order to examine the performance of the proposed incremental construction methods of fuzzy rule-based classifiers using a simple artificial pattern classification problem.