Study on algorithms of determining basic probability assignment function in Dempster-Shafer evidence theory

Abstract

Dempster-Shafer evidence theory (D-S theory) provides a useful computational scheme for integrating uncertainty information from multiple sources in artificial intelligence systems. D-S evidence theory is a useful method for dealing with uncertainty problems. Therefore, it has been successfully applied in data fusion and pattern recognition. However, it also has some shortcomings. The key problem to D-S reasoning is basic probability assignment (BPA) function, which to a great extent limits its applications. To solve this problem, this paper presents three methods to constructing the BPA function. These methods are based on gray correlation analysis, fuzzy sets, and attribute measure respectively. Furthermore, experiments of recognizing the emitter purpose are selected to demonstrate these methods of determining the BPA function proposed. Experimental results show that the performance of these new methods is accurate and effective.