HP Labs Technical Reports

A Type 2 Fuzzy Set Based Model for Adaptive Information Retrieval

Piccinelli, Giacomo; Casassa Mont, Marco

HPL-98-27

Keyword(s): fuzzy set; information retrieval; adaptivity; DBMS

Abstract: The abstraction model that an information retrieval (IR) system imposes on data determines its knowledge on them. Uncertainty permeates the behaviour of both the system and the users and we argue that its explicit modelling is fundamental in order to increase classical IR parameters like precision and recall as well as to reduce the gap between systems and users. We present a keyword based model that, capitalising on the flexibility of fuzzy sets, extends the traditional two dimensional vector approach to data abstraction, evolving it into a paradigm where relevance is tightly coupled with uncertainty and the view the system has on data evolves dynamically through an adaptivity process. A prototype system (DUNE) has been derived from the general model and we are currently evaluating possible applications of our proposal in the knowledge management field.

11 Pages

Back to Index