Comparing and Combining Methods for Automatic Query Expansion

TL;DR

This paper compares cooccurrence and probabilistic methods for automatic query expansion, demonstrating that combining these approaches yields better retrieval performance than using either method alone.

Contribution

It introduces a naive combination of cooccurrence and probabilistic query expansion methods, showing improved results over individual approaches.

Findings

Combined methods outperform individual approaches in query expansion.

Different approaches provide complementary information for query expansion.

Results confirm the benefit of integrating multiple query expansion techniques.

Abstract

Query expansion is a well known method to improve the performance of information retrieval systems. In this work we have tested different approaches to extract the candidate query terms from the top ranked documents returned by the first-pass retrieval. One of them is the cooccurrence approach, based on measures of cooccurrence of the candidate and the query terms in the retrieved documents. The other one, the probabilistic approach, is based on the probability distribution of terms in the collection and in the top ranked set. We compare the retrieval improvement achieved by expanding the query with terms obtained with different methods belonging to both approaches. Besides, we have developed a na\"ive combination of both kinds of method, with which we have obtained results that improve those obtained with any of them separately. This result confirms that the information provided by each approach is of a different nature and, therefore, can be used in a combined manner.