Concept-based document models using explicit semantic analysis

Explicit semantic analysis (ESA) is a novel method that represents the meaning of texts in a high-dimensional space of concepts derived from Wikipedia. Several different ESA-based retrieval models have been proposed. However, these approaches depended on the performance of pseudo relevance feedback. In this paper, we propose a concept-based retrieval model under the language modeling framework. By means of concept mapping using explicit semantic analysis, the original documents are translated into conceptual representations, which are subsequently used to update the document models. The concept-based document model is evaluated on the TREC Ad Hoc Track (Disks 1, 2, and 3) collections. Experiment results show significant improvements with respect to the baseline models.