Robust Combining of Disparate Classi ers through Order Statistics

Integrating the outputs of multiple classiiers via combiners or meta-learners has led to substantial improvements in several diicult pattern recognition problems. In this article we investigate a family of combiners based on order statistics, for robust handling of situations where there are large discrepancies in performance of individual classiiers. Based on a mathematical modeling of how the decision boundaries are aaected by order statistic combiners, we derive expressions for the reductions in error expected when simple output combination methods based on the the median, the maximum and in general, the i th order statistic, are used. Furthermore, we analyze the trim and spread combiners, both based on linear combinations of the ordered classiier outputs, and show that in the presence of uneven classiier performance, they often provide substantial gains over both linear and simple order statistics combiners. Experimental results on both real world data and standard public domain data sets corroborate these ndings.