论文信息 - Question Answering System: Retrieving Relevant Passages

Question Answering System: Retrieving Relevant Passages

This paper discusses the QA system submitted by Dhirubhai Ambani Institute of Information and Communication Technology, India in the ResPubliQA 2010. We have participated in the monolingual en-en task. Our system retrieves a candidate paragraph that contains the answer to a natural language question. Depending on the n-gram similarity score of the candidate paragraph, a decision is made whether to answer the question or not. The objective of our participation is to test our implementation of various strategies like Query Expansion, n-gram similarity matching, and non-answering criteria.

Prasenjit Majumder | Hitesh Sabnani | Prasenjit Majumder | Hitesh Sabnani

[1] Anselmo Peñas,et al. Overview of ResPubliQA 2009: Question Answering Evaluation over European Legislation , 2009, CLEF.

[2] Paolo Rosso,et al. NLEL-MAAT at CLEF-ResPubliQA , 2009, CLEF.

[3] Jimmy J. Lin,et al. Overview of the TREC 2007 Question Answering Track , 2008, TREC.

[4] Marius Paca. Open-Domain Question Answering from Large Text Collections , 2003, Computational Linguistics.