Title: An intelligent system for sentence retrieval and novelty mining

Authors: Flora S. Tsai, Kap Luk Chan

Addresses: Nanyang Technological University, School of Electrical and Electronic Engineering, 639798, Singapore. ' Nanyang Technological University, School of Electrical and Electronic Engineering, 639798, Singapore

Abstract: This paper describes the development of an intelligent system for sentence retrieval and novelty mining. Knowledge and software engineering techniques are used for the development, implementation and evaluation strategies for relevant sentence retrieval and novelty mining using blended metrics. Moreover, we describe our novelty mining system with result aggregation, which can greatly facilitate users to find novel information. The experimental results on TREC 2003 and TREC 2004 novelty track data demonstrate the usefulness of our intelligent novelty mining system in a variety of performance situations and user settings. By considering the special issues based on domain knowledge, our intelligent novelty mining system can help to discover novelty that can support business people who are making domain-pertinent actionable decisions.

Keywords: intelligent systems; sentence retrieval; novelty mining; evaluation measure; blended metrics; knowledge engineering; software engineering; novel information; relevant information; information retrieval.

DOI: 10.1504/IJKEDM.2011.037645

International Journal of Knowledge Engineering and Data Mining, 2011 Vol.1 No.3, pp.235 - 253

Published online: 07 Mar 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article