Soft combination of neural classifiers: A comparative study

This paper presents four schemes for soft fusion of the outputs of multiple classifiers. In the first three approaches, the weights assigned to the cl assifiers or groups of them are data dependent. The first approach involves the calculation of fuzzy integrals. The second scheme performs weighted av eraging with data-dependent weights. The third approach performs linear com bination of the outputs of classifiers via the BADD defuzzification strateg y. In the last scheme, the outputs of multiple classifiers are combined usi ng Zimmermann's compensatory operator. An empirical evaluation using widely accessible data sets substantiates the validity of the approaches with dat a-dependent weights, compared to various existing combination schemes of mu ltiple classifiers. (C) 1999 Elsevier Science B.V. All rights reserved.