Semi-fuzzy Quantifiers for Information Retrieval

Recent research on fuzzy quantification for information retrieval has proposed the application of semi-fuzzy quantifiers for improving query languages. Fuzzy quantified sentences are useful as they allow additional restrictions to be imposed on the retrieval process unlike more popular retrieval approaches, which lack the facility to accurately express information needs. For instance, fuzzy quantification supplies a variety of methods for combining query terms whereas extended boolean models can only handle extended boolean-like operators to connect query terms. Although some experiments validating these advantages have been reported in recent works, a comparison against state-of-the-art techniques has not been addressed. In this work we provide empirical evidence on the adequacy of fuzzy quantifiers to enhance information retrieval systems. We show that our fuzzy approach is competitive with respect to models such as the vector-space model with pivoted document-length normalization, which is at the heart of some high-performance web search systems. These empirical results strengthen previous theoretical works that suggested fuzzy quantification as an appropriate technique for modeling information needs. In this respect, we demonstrate here the connection between the retrieval framework based on the concept of semi-fuzzy quantifier and the seminal proposals for modeling linguistic statements through Ordered Weighted Averaging operators (OWA).