A Model for Generating Related Weighted Boolean Queries
In this paper a model for polyrepresenting weighted boolean queries based on reformulation is presented. This model tries to generate a set of boolean queries for completing the meaning of the original query instead of modifying the query adding, changing or removing terms as techniques such as query expansion propose. This model is specially designed for those systems which do not support very large queries such as the Web search engines.
The input of the model are boolean queries which have been weighted by the user in order to express some semantics that the set of polyrepresented queries have to satisfy. The user is able to fix a threshold for limiting the minimum resemblance of the polyrepresented queries and the original one. From the original query, using a reformulation strategy which guarantees that length of the new queries is not enlarged excessively, the proposed model generates a set of candidate queries/concepts for representing the original one. That set is reduced by a bottom-up process based on the semantics established by the user in order to choose the best queries for representing the original one.
KeywordsQuery polyrepresentation Weighted boolean queries Fuzzy logic
Unable to display preview. Download preview PDF.
- 3.Bruza, P., Dennis, S.: Query Reformulation on the Internet: Empirical Data and the Hyperindex Search Engine. In: Proceedings of RIAO ’97 - Computer Assisted Information Searching on the Internet, pp. 500–509 (1997)Google Scholar
- 4.Efthimiadis, E.N.: A user-centred evaluation of ranking algorithms for interactive query expansion. In: SIGIR ’93: Proceedings of the 16th annual international ACM SIGIR conference, New York, NY, USA, pp. 146–159 (1993)Google Scholar