Abstract | ||
---|---|---|
In this paper, we present two different question-answering systems on speech transcripts which participated to the QAst 2007 evaluation. These two systems are based on a complete and multi-level analysis of both queries and documents. The first system uses handcrafted rules for small text fragments (snippet) selection and answer extraction. The second one replaces the handcrafting with an automatically generated research descriptor. A score based on those descriptors is used to select documents and snippets. The extraction and scoring of candidate answers is based on proximity measurements within the research descriptor elements and a number of secondary factors. The evaluation results are ranged from 17% to 39% as accuracy depending on the tasks. |
Year | DOI | Venue |
---|---|---|
2007 | 10.1007/978-3-540-85760-0_55 | Advances in Multilingual and Multimodal Information Retrieval |
Keywords | DocType | Volume |
candidate answer,handcrafted rule,secondary factor,proximity measurement,multi-level analysis,evaluation result,answer extraction,qast track,research descriptor element,different question-answering system,research descriptor,limsi participation,question answering system | Conference | 5152 |
ISSN | Citations | PageRank |
0302-9743 | 14 | 1.14 |
References | Authors | |
6 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Sophie Rosset | 1 | 393 | 61.66 |
Olivier Galibert | 2 | 314 | 30.03 |
Gilles Adda | 3 | 1016 | 137.10 |
Eric Bilinski | 4 | 57 | 9.39 |