Content-Based Mining of Query Replacements

This dissertation introduces the problem of content-based query replacement. It motivatesthe term correlation modeled using MRFs and the sampling based technique to learn thesearch phrase definitions. These definitions are used as alternative queries to achieve higheraccuracy in several retrieval tasks.Retrieving relevant documents while eliminating irrelevant documents for an user'squery is a challenging task which involves a good understanding of the relation betweenthe data and the query as well as developing algorithms that can efficiently measure therelevance of the data to the query. As part of this dissertation, we have developed a hypothesis to reduce the problem of mining search phrase definitions significantlyby modeling the joint distribution of terms as a product of conditional distributions,modeled as a Markov Random Field. We assume that there exists an underlying jointdistribution among terms that are used to describe the search phrase. The modelingwe propose is a condensed representation of inter-term relationship and it appears tocapture insight statistics among terms. describes a target phrase.