to improve the accuracy of answering the question series of the gathering type, on the other hand, Strategy II is effective for the question series of the browsing type.

-EXPANSION. They have been taxonomized as the Webclopedia QA Typology (Hermjakob el al. 2002). The system increasingly makes use of syntactic and semantic (world) knowledge to improve the accuracy of its results. We will identify the strength and weakness of our approach during the demonstration with examples. Figure

Elementary-level science exams pose sig-nificant knowledge acquisition and rea-soning challenges for automatic question answering. We develop a system that rea-sons with knowledge derived from text-books, represented in a subset of first-order logic. Automatic extraction, while scalable, often

enables responses drawn from generated/new knowledge, and (c) key phrase based search when (a) and (b) are not possible. The design allows for two levels of response accuracy improvement over standard search, while retaining a minimum performance level of standard search capabilities. 1

a simple bag-ofwords representation has been small. We show that most questions reveal their class through a short contiguous token subsequence, which we call its informer span. Perfect knowledge of informer spans can enhance accuracy from 79.4 % to 88% using linear SVMs on standard benchmarks

of analyses of the in-dividual translations. Finally, we present a language-augmented method that enriches the questions and answers with paraphrases obtained by means of machine translation. We show that all of the described ap-proaches achieve a significant improvement over the random baseline

tells them that they are unlikely to be caught A wide range of predictions are also possible for those viewing the direct and strategic interrogative questions and answers. Clearly, previous research leads to a prediction of slightly above-chance accuracy. To the extent that previous findings

Community Question Answering (CQA) services enable their users to exchange knowledge in the form of questions and answers. These communities thrive as a result of a small number of highly active users, typically called experts, who provide a large number of high quality useful answers. Expert

the potential of automated reasoning. Performances in the recent TREC-QA tracks show that inferencing substantially improves the response relevance and accuracy. There is still a long way to go before we can consider our document repositories (such as the World Wide Web) as a huge knowledge base with question

This paper describes several experiments combining natural language and acoustic constraints to improve overall performance of the MIT VOYAGER spoken language system. This system cou-ples the SUMMIT speech recognition system with the TINA lan-guage understanding system to answer spoken queries