Integration of association rules and ontology for semantic-based query expansion

Paper accepted for publication in Data and Knowledge Engineering. Retrieved 6/26/2006 from http://www.ischool.drexel.edu/faculty/thu/My%20Publication/Journal-papers/song_dek_2006.pdf.The goal of query expansion is to reduce the mismatch between documents and queries by expanding the query using words or phrases with a similar meaning or some other statistical relation to the set of relevant documents. One of the limitations with query expansion techniques is that a query is often expanded only by the linguistic features of terms. To tackle this problem, we propose a novel semantic query expansion technique that combines association rules with ontologies and Natural Language Processing techniques. Our technique utilizes the association rule discovery to find good candidate terms to improve the retrieval performance. These candidate terms are automatically derived from collections and added to the original query. Our technique is differentiated from others in that 1) it utilizes the semantics as well as linguistic properties of unstructured text corpus, 2) it makes use of contextual properties of important terms discovered by association rules, and 3) ontologies’ entry is added to the query by disambiguating word senses. Experiments conducted on TREC collections give encouraging results. We achieve from 13.41% to 32.39% improvement in term of P@20 and from 8.39% to 14.22% in terms of F-measure with TREC ad hoc queries. Detailed descriptions of the experimental results are discussed in the paper