Yasuhiko Watanabe
ABSTRACT
The most serious difficulty in developing a QA system is knowledge. In this paper we first discuss three problems of developing a knowledge base by which a QA system answers how type questions. Then we propose a method of developing a knowledge base by using mails posted to a mailing list. Next we describe a QA system which can answer how type questions based on the knowledge base. Our system finds question mails which are similar to user's question and shows the answers to the user. The similarity between user's question and a question mail is calculated by matching of user's question and a significant sentence in the question mail. Finally we show that mails posted to a mailing list can be used as a knowledge base by which a QA system answers how type questions.
- TREC (Text REtrieval Conference): http://trec.nist.gov/Google Scholar
- NTCIR (NII-NACSIS Test Collection for IR Systems) project: http://research.nii.ac.jp/ntcir/index-en.htmlGoogle Scholar
- Kurohashi and Higasa: Dialogue Helpsystem based on Flexible Matching of User Query with Natural Language Knowledge Base 1st ACL SIGdial Workshop on Discourse and Dialogue pp.141--149 (2000). Google ScholarDigital Library
- Kiyota Kurohashi and Kido: "Dialog Navigator" A Question Answering System based on Large Text Knowledge Base 19th COLING (COLING02) pp.460--466 (2002.8). Google ScholarDigital Library
- Kurohashi and Nagao: A syntactic analysis method of long Japanese sentences based on the detection of conjunctive structures Computational Linguistics 20(4)pp.507--534 (1994). Google ScholarDigital Library
- Kurohashi and Nagao: JUMAN Manual version 3.6 (in Japanese) Nagao Lab. Kyoto University (1998).Google Scholar
Index Terms
- A question answer system using mails posted to a mailing list
Recommendations
A question answer system using mails posted to a mailing list
IMTCI'04: Proceedings of the Second international conference on Intelligent Media Technology for Communicative IntelligenceThe most serious difficulty in developing a QA system is a lack of knowledge. In this paper, we first discuss three problems of developing a knowledge base by which a QA system answers How-type questions. Then, we propose a method of developing a ...
Invalidation of Mailing List Address to Block Spam Mails
APSCC '08: Proceedings of the 2008 IEEE Asia-Pacific Services Computing ConferenceMailing lists are used for information exchange in specific groups. However, in the recent times, the number of spam mails received has increased, and considerable amount of time is wasted in filtering spam mails. Spam filtering techniques are widerly ...
Web-based mailing list administration on RMX
iiWAS '15: Proceedings of the 17th International Conference on Information Integration and Web-based Applications & ServicesThe Rule-based e-Mail eXchange system, also known as RMX, is a mail transfer agent that transfers emails based on user-defined rules, which are applied to a relational database. It is a novel way of sending email to various groups of people, compared to ...
Reviews
Hans-Christian Schmitz
A system for answering "How__?__" type questions is described in this paper. The system consists of a Web-based user interface, a processor for analyzing Japanese input questions and determining output answers, and a knowledge base. The knowledge base is filled with email messages that were posted to a mailing list. From each message, a 'significant' sentence is extracted. Among the clues for determining such a sentence are occurrences of nouns that also occur in the subject line, occurrences of expressions that typically occur in significant sentences, and quotations of the sentence. The messages from the mailing list stand in rhetorical relation to each other: there are messages with new questions; messages that directly respond to new questions, for example by giving an answer or posing a clarification question; and messages with questioner's replies, for example, responses to answers or clarification questions. If two messages stand in a rhetorical relation, then their significant sentences are also related. If a user enters an input question to the system, then the similarity scores of the input question and the significant sentences from the knowledge base are calculated. The system finds a sentence that is most similar to the input question; it outputs this sentence, together with its rhetorically related sentences. The system approach is interesting. The user receives an extract of a whole email conversation as an answer to his or her question. This allows the user to compensate for problems connected with wrong answers (usually followed by a correction), and imperfect questions (usually followed by a clarification). Overall, this paper is well structured, and very clear. Online Computing Reviews Service
Access critical reviews of Computing literature here
Become a reviewer for Computing Reviews.
Comments