| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To disclose methods of providing a document relevance score to a document on a network. <P>SOLUTION: A query-independent usage parameter QID with a document usage value U considered is generated to each of a plurality of documents on a network, the usage value U is the total time or an average time that all users use for a document within a given time period if there is actual usage data, the usage value U is a default usage value of if there are no actual usage data, the plurality of documents are searched on the network by a search query including a search string input by a user to generate a relevance score QD representing a relevance with the search string of the plurality of documents, and a ranking function for calculating a ranking score is used on the basis of the query-independent usage parameter QID corresponding to each document and the relevance score QD to rank the plurality of documents. <P>COPYRIGHT: (C)2011,JPO&INPIT</p> |
