State-of-the-art question answering (QA) systems employ termdensity ranking to retrieve answer passages. Such methods often retrieve incorrect passages as relationships among question terms are not considered. Previous studies attempted to address this problem by matching dependency relations between questions and answers. They used strict matching, which fails when semantically equivalent relationships are phrased differently. We propose fuzzy relation matching based on statistical models. We present two methods for learning relation mapping scores from past QA pairs: one based on mutual information and the other on expectation maximization. Experimental results show that our method significantly outperforms state-of-the-art density-based passage retrieval methods by up to 78 % in mean reciprocal rank. Relation matching also brings about a 50 % improvement in a system enhanced by query expansion.

Work on the semantics of questions has argued that the relation between a question and its answer(s) can be cast in terms of logical entailment. In this paper, we demonstrate how computational systems designed to recognize textual entailment can be used to enhance the accuracy of current open-domain automatic question answering (Q/A) systems. In our experiments, we show that when textual entailment information is used to either filter or rank answers returned by a Q/A system, accuracy can be increased by as much as 20 % overall. 1

We present an approach to query expansion in answer retrieval that uses Statistical Machine Translation (SMT) techniques to bridge the lexical gap between questions and answers. SMT-based query expansion is done by i) using a full-sentence paraphraser to introduce synonyms in context of the entire query, and ii) by translating query terms into answer terms using a full-sentence SMT model trained on question-answer pairs. We evaluate these global, context-aware query expansion techniques on tfidf retrieval from 10 million question-answer pairs extracted from FAQ pages. Experimental results show that SMTbased expansion improves retrieval performance over local expansion and over retrieval without expansion. 1

This paper presents a syntax-driven approach to question answering, specifically the answer-sentence selection problem for short-answer questions. Rather than using syntactic features to augment existing statistical classifiers (as in previous work), we build on the idea that questions and their (correct) answers relate to each other via loose but predictable syntactic transformations. We propose a probabilistic quasi-synchronous grammar, inspired by one proposed for machine translation (D. Smith and Eisner, 2006), and parameterized by mixtures of a robust nonlexical syntax/alignment model with a(n optional) lexical-semantics-driven log-linear model. Our model learns soft alignments as a hidden variable in discriminative training. Experimental results using the TREC dataset are shown to significantly outperform strong state-of-the-art baselines. 1

This work describes an answer ranking engine for non-factoid questions built using a large online community-generated question-answer collection (Yahoo! Answers). We show how such collections may be used to effectively set up large supervised learning experiments. Furthermore we investigate a wide range of feature types, some exploiting NLP processors, and demonstrate that using them in combination leads to considerable improvements in accuracy. 1

At TREC-2003, TextMap participated in the Main task, which encompassed answering the following types of questions: • factoid questions; • list questions;

Many question-answering systems start with a passage retrieval system to facilitate the answer extraction process. The richer the set of passages, in terms of answer content, the more accurate the answer extraction. We present a simple translation model for passage retrieval at the sentence level. We demonstrate this framework on TREC data, and show that it performs better than retrieval based on query likelihood, and on par with other systems. 1.

In this paper we describe Tokyo Institute of Technology’s submission to the TAC2008 question answering (QA) track. Keeping the same theoretical QA model as for the TREC2007 track, developed for factoid QA, we investigated the effects of retrieving blog data versus web data for rigid list questions. For squishy list questions we relied on sentence retrieval, similar to how we approached “other ” questions in TREC2007. While our performance on rigid list questions was poor, for squishy list questions we achieved a score only slightly lower than the highest score of all participants. 1.

Progress in Question Answering can be achieved by (1) combining multiple strategies that optimally resolve different question classes of various degrees of complexity; (2) enhancing the precision of question interpretation and answer extraction; and (3) question decomposition and answer fusion. In this paper we also present the impact of modeling the user background on Q/A and discuss the pragmatics pf processing negation in Q/A. 1

www.lti.cs.cmu.edu