| 摘要 |
[Object] To provide a question-answering system capable of appropriately answering to a non-factoid question.;[Solution];A question-answering system 30 includes: a related document searching unit 54 responsive to a question, for taking out answer candidates from an object document storage unit 32; an answer candidate extracting unit 56 extracting plausible ones from the answer candidates; a causal relation recognizing unit 60 recognizing causal relation expression included in extracted answer candidates; causal relation relevancy determining unit 64 determining whether or not a recognized causal relation is relevant as an answer; a feature vector generating unit 66 generating a feature vector related to a combination of the question and the searched answer candidate using the results of processing up to the causal relation relevancy determining unit 64; and a feature vector generating unit 66 for calculating, when a feature vector is given, a score indicating a degree of how correct an answer candidate is as an answer to the question, for the question and the answer candidate as a source of the feature vector, and for outputting, as the answer to the question, an answer candidate having the highest score. |
| 主权项 |
1. A non-factoid question-answering system, receiving a non-factoid question and generating an answer to the question, used connected to document storage means for storing a plurality of computer-readable documents, comprising:
candidate search means, responsive to an input of a question, for taking out answer candidates to the question from said document storage means; feature generation means, responsive to said candidate search means having searched answer candidates, for generating, for combinations of said question and each of the answer candidates searched by said candidate search means, a set of prescribed features; and answer selection means for calculating, when a set of said features generated by said feature generation means is given, for the question and the answer candidate as a base for generating the set of features, a score representing a degree of plausibility of the answer candidate as a correct answer to the question, and outputting an answer candidate having the highest score as an answer to the question; wherein said feature generation means includes first feature calculating means for calculating, for said question and each answer candidate, a first set of features obtained from respective morpheme information and syntactic information, and second feature calculating means for selecting a causal relation expression in each of said answer candidates relevant as giving an answer to said question, and for calculating a second set of features obtained from said causal relation expression; and said set of features include said first set of features and said second set of features. |
