Information retrieval estimation via fuzzy probability

Abstract

Fuzzy logic has been recognized as a useful approach for information retrieval. It helps identify partially matched documents for a given query so that ranking the relevant documents becomes straightforward. Unfortunately, research on fuzzy information retrieval only focuses on the estimation of relevancy of the retrieved documents, but lacks the estimation of imprecise probability of obtaining such relevant documents. This paper makes an attempt to the latter. It extends the fuzzy probability (FP) calculation method proposed by Huang and Shi to an enhanced FP calculation method, Furthermore, it proposes a novel FP calculation method. An example is provided to illustrate the use of fuzzy probability estimation.