A study on query modification methods for web search using taxonomy and classification learning タキソノミーと分類学習を用いたWeb探索のための問合せ修正法に関する研究

博士論文

A pressing need exists to improve effectiveness of the web search process and result. The need results from the extremely large volume of information available on the web and the large variety of Internet users. Current web search services are the main startig points to search for information needed from the huge document collection. Their limitations, however, are apparent. Search engines that employ a keyword-based search are good at returning a long list of relevant documents, but the list is also cluttered with many irrelevant candidates. Even if the engines rank the matched documents according to a specific ranking algorithm, they are still not especially helpful in locating the needed information. This problem results mainly from the short queries usually given by users, where it is hard to express the search intent using short queries and the keyword search approach used by the search engines. ･･･

2002

Includes bibliographical references