Contextual paradigm for ad hoc retrieval of user-centric web data

Abstract

The choice of the term-weighting technique and the combination of term weights during retrieval is an important aspect of the web information retrieval (WebIR) system. At their heart, most WebIR models utilise some form of term frequency-based term-weighting technique. The notion is that the more often a query term occurs in a document, the more likely it is that the document meets an information need. We examine an alternative. We propose a model, which assesses the presence of a term in a document not by looking at the actual occurrence of that term, but by a set of non-independent supporting terms, that is context. We expound contextual proximity model (CPM), a novel context-based paradigm, which when used as a part of the web retrieval process, will improve retrieval effectiveness in response to a given multi-term input query. This work is concerned with a non-iterative retrieval process, that is one without query refinement or relevance feedback.