Pattern classification using multiple hierarchical overlapped self-organising maps

Abstract In this paper, we describe techniques for designing high-performance pattern classification systems using multiple hierarchical overlapped self-organising maps (HOSOM) (Suganthan, Proceedings of the International Joint Conference on Neural Networks, WCCI’98, Alaska, 1998). The HOSOM model has one first level SOM and several partially overlapping second-level SOMs. With this overlap, every training and test sample is classified by multiple second-level SOMs. Hence, the final classification decision can be made by combining these multiple classification decisions to obtain a better performance. In this paper, we use multiple HOSOMs and each HOSOM is trained on a distinct input feature set extracted from the same data set. Since one HOSOM yields multiple classifications, these multiple HOSOMs generate a large number of classification decisions. To combine the individual classifications, we make use of the global winner as well as a winner for every class. Our experiments yielded a high recognition rate of 99.25% on NIST19 numeral database.