Automatic acquisition of hyponyms from large text corpora
COLING '92 Proceedings of the 14th conference on Computational linguistics - Volume 2
Finding specification pages according to attributes
Proceedings of the 15th international conference on World Wide Web
Proceedings of the 16th international conference on World Wide Web
Automatic discovery of attribute words from web documents
IJCNLP'05 Proceedings of the Second international joint conference on Natural Language Processing
Hi-index | 0.00 |
This paper presents a novel web-oriented writing environment that helps users describe their opinions on topics/events through weblogs, by showing facts related to the objects. Our writing environment automatically recognizes information needs for facts on aspects (e.g., "opening hours'') of the objects (e.g., "J's cafe'') by applying lexico-syntactic patterns to the snapshots of users' writings. It then attempts to extract facts that satisfy the information needs from the Web, and shows the extracted facts (e.g., "opening hours''-"10am-8:30pm'') to the users. The users can thus refer to necessary facts without searching the web by themselves. We realized this environment for Japanese on a browser with AJAX, and evaluated how much useful information could be provided by our environment.