<XML><RECORDS><RECORD><REFERENCE_TYPE>1</REFERENCE_TYPE><REFNUM>8434</REFNUM><AUTHORS><AUTHOR>Azzopardi,L.</AUTHOR></AUTHORS><YEAR>2005</YEAR><TITLE>Incorporating Context within the Language Modeling Approach for ad hoc Information Retrieval</TITLE><PLACE_PUBLISHED>PhD Thesis, University of Paisley</PLACE_PUBLISHED><PUBLISHER>N/A</PUBLISHER><LABEL>Azzopardi:2005:8434</LABEL><ABSTRACT>In this thesis, we investigate using the Language Modeling approach for ad hoc Information Retrieval as a theoretically principled framework for encoding contextual evidence. Using context to improve retrieval performance is a current challenge within the discipline and presents a major challenge to the research community. The Language Modeling approach provides a natural and intuitive means of encoding the context associated with a document. However, the Language Modeling approach also represents a change to the way probability theory is applied in ad hoc Information Retrieval and makes several assumptions for its application[112, 113, 57, 96]. We consider these assumptions and study them in detail during the course of this thesis. Central to the assumptions is the key implication that better retrieval performance can be obtained through developing better representation of the documents. We posit that the context associated with a document will enable the development of such representations - context based document models. This premise relies upon the explicit and implicit assumptions of the Language Modeling approach being valid, which have, up until now, not been fully tested or verified. Through the course of this thesis we (1) formalize the assumptions of the Language Modeling approach; (2) motivated by the implications of these assumptions we present our framework for estimating context based document models; (3) perform a comprehensive analysis of the main assumptions underlying the Language Modeling approach, not only to validate the approach, but to deepen our understanding of the retrieval model itself, and; (4) empirically assess the performance of the context based document models against the standard document models on various test collections and contexts. Our findings show that there are occasions when context based document models outperform the standard document model. Further analysis with respect to underlying assumptions though reveals some of the limitations of the Language Modeling approach. We discuss these limitations and suggest an alternative approach for embedding context within the model. Finally, we propose an Integrated Language Modeling approach which formalizes the existing theory and practice within one framework. This not only addresses some of the concerns over the standard Language Modeling approach, but also enables the integration of various forms of context within the one framework.</ABSTRACT></RECORD></RECORDS></XML>