Bidirectional Approximate Reasoning Based on Weighted Similarity Measures of Vague Set

In this paper, we introduce a new weighted similarity measure for vague set in order to solve some faults in [1] , and propose the concept of similarity direction between two vague sets by which to describe which one could give more accurate information. At the same time, we propose a method for determining the similarity direction. Based on above, we present a bidirectional approximate reasoning method based on weighted similarity measures of vague set, which fully uses the message accuracy of vague set. Then this method improves the accuracy and applicability of approximate reasoning, and also provides a useful tool for approximate reasoning in intelligence system.