Abstract | ||
---|---|---|
We propose a method of discovering alternative answers from the Web that are to a question posted on a Web question & answer (Q&A) site and differ from existing answers to the question on the Q&A site. Our method first automatically generates queries for conventional Web search engines to collect Web pages that can contain answers to the target question. Each collected Web page is evaluated by calculating two kinds of scores: one represents the probability that the page has information that answers a question in the Q&A content and the other represents the possibility that it has an alternative answer. The method is implemented and the system is evaluated using actual Q&A contents. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1007/978-3-642-14246-8_43 | WAIM |
Keywords | Field | DocType |
target question,alternative answer,actual q,web question,webqa site,conventional web search engine,web page,web search engine,question answering,web pages | Web search engine,Data mining,World Wide Web,Site map,Search engine,Information retrieval,Web page,Computer science | Conference |
Volume | ISSN | ISBN |
6184 | 0302-9743 | 3-642-14245-1 |
Citations | PageRank | References |
1 | 0.41 | 11 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Natsuki Takata | 1 | 1 | 0.41 |
Hiroaki Ohshima | 2 | 236 | 28.19 |
Satoshi Oyama | 3 | 265 | 34.67 |
Katsumi Tanaka | 4 | 1349 | 160.89 |