Evaluating score normalization methods in data fusion

Abstract

In data fusion, score normalization is a step to make scores, which are obtained from different component systems for all documents, comparable to each other. It is an indispensable step for effective data fusion algorithms such as CombSum and CombMNZ to combine them. In this paper, we evaluate four linear score normalization methods, namely the fitting method, Zero-one, Sum, and ZMUV, through extensive experiments. The experimental results show that the fitting method and Zero-one appear to be the two leading methods. © Springer-Verlag Berlin Heidelberg 2006.