[Show abstract][Hide abstract] ABSTRACT: This paper describes FERRET, an interactive question-answering (Q/A) system designed to address the challenges of integrating automatic Q/A applications into real-world environments. FERRET utilizes a novel approach to Q/A - known as predictive questioning - which attempts to identify the questions (and answers) that users need by analyzing how a user interacts with a system while gathering information related to a particular scenario.
ACL 2006, 21st International Conference on Computational Linguistics and 44th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference, Sydney, Australia, 17-21 July 2006; 01/2006
[Show abstract][Hide abstract] ABSTRACT: This paper describes a novel framework for interactive question-answering (Q/A) based on predictive questioning. Gen- erated off-line from topic representations of complex scenarios, predictive ques- tions represent requests for information that capture the most salient (and diverse) aspects of a topic. We present experimen- tal results from large user studies (featur- ing a fully-implemented interactive Q/A system named FERRET) that demonstrates that surprising performance is achieved by integrating predictive questions into the context of a Q/A dialogue.
ACL 2005, 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference, 25-30 June 2005, University of Michigan, USA; 01/2005
[Show abstract][Hide abstract] ABSTRACT: In 2004, LCC has participated to TREC QA with two systems; PowerAnswer 2 and PALANTIR. The goal was to study the contri-bution of new features such as: (1) an improved passage retrieval module based on the Lucene IR system, (2) multiple QA strategies with vot-ing across strategies, (3) the use of text seman-tic relations, (4) a new method for extracting lexical chains, (5) new logic prover reasoning strategies, (6) anaphora and ellipsis resolution, (7) indexing based on named entities, (8) ques-tion similarity analysis, and others. The paper presents the architectures of these two systems, an error analysis, and summary of results.
[Show abstract][Hide abstract] ABSTRACT: This paper addresses the pragmatic challenges that state-of-the-art question/answering sys- tems face in trying to decompose complex information-seeking scenarios. We propose that question decomposition can be approached in one of two ways: either by approximating the domain-specific knowledge for a particular set of domains, or by identifying the decompo- sition strategies employed by human users. We also present preliminary results from experi- ments that confirm the viability of each of these approaches within an interactive Q/A context.
[Show abstract][Hide abstract] ABSTRACT: This paper describes the development of CICEROARABIC, the first wide coverage named entity recognition (NER) system for Modern Standard Arabic. Capable of classifying 18 different named entity classes with over 85% F, CICEROARABIC utilizes a new 800,000-word annotated Arabic newswire corpus in order to achieve high performance without the need for hand-crafted rules or morphological information. In addition to describing results from our system, we show that accurate named entity annotation for a large number of semantic classes is feasible, even for very large corpora, and we discuss new techniques designed to boost agreement and consistency among annotators over a long-term annotation effort.