Task-Centric Document Recommendation via Context-Specific Terms

Surendra Sarnikar, J. Leon Zhao

Context-specific document recommender systems rely on the accurate identification of context descriptors from unstructured textual information to identify highly relevant documents. In this paper, we propose two term-weighting measures, normal distance and adjusted inverse polysemy, to enable the retrieval of relevant documents with higher precision. We analyze the performance of the proposed measures and present results with respect to a domain-specific corpus.

Subjects: 1.10 Information Retrieval; 13. Natural Language Processing

Submitted: May 17, 2006

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.