A genetic algorithm for determining nonadditive set functions in information fusion

As a classical aggregation tool, the weighted average method is widely used in information fusion. It is the Lebesgue integral with respect to the wei ghts essentially. Due to some inherent interaction among diverse informatio n sources, the weighted average method does not work well in many real prob lems. To describe the interaction, an intuitive and effective way is to rep lace the additive weights with a nonadditive set function defined on the po wer set of the set of all information sources. Instead of the weighted aver age method, we should use the Choquet integral or some other nonlinear inte grals, especially, the new nonlinear integral introduced by the authors rec ently. The crux of making such an improvement is how to determine the nonad ditive set function from given input-output data when the nonlinear integra l is viewed as a multiinput single-output system. In this paper, we employ a specially designed genetic algorithm to realize the optimization in deter mining the nonadditive set function. (C) 1999 Elsevier Science B.V. All rig hts reserved.