Application of Fuzzy Set Theory to Extend Boolean Information Retrieval
Gloria Bordogna, Gabriella Pasi · Studies in fuzziness and soft computing · 2000
Abstract. The primary objective of the extensions of Boolean information re-trieval within fuzzy set theory is to model the graduality of the concept of relevance of documents to a user’s query. The result of a query evaluation is represented as a fuzzy subset of the archived documents. Several fuzzy extensions of the Boolean model have been defined which share the characteristics of adopting a weighted document representation and a weighted query language: while these extensions retain the same semantics for the index term weights, named significance degrees, nevertheless, they differ in the semantics associated with the query weights. Query weights are introduced as attributes of the search terms to provide for a greater expressiveness in the formulation of the information needs. The aim of this con-tribution is to show how the fuzzy Boolean information retrieval models are more flexible in representing both document contents and information needs; this char-acteristics is provided by their ability to represent and manage linguistic concepts having a gradual nature.