| 主权项 |
1. A computer-implemented process for retrieving online information for a user, comprising:
using a computer to perform the following process actions: receiving an original query submitted by the user, the original query comprising a plurality of query terms that the user associates with particular online information the user is interested in; analyzing the original query to determine central concepts of the original query, said analysis comprising,
segmenting the original query by extracting main concepts therefrom, each extracted main concept comprising one or more of the query terms, and then,retrieving queries from a historical query log of a search engine which are hypothesized to be semantically related to the main concepts extracted from the segmented original query, said historical query log comprising said queries which were received from a collection of search engine users,extracting salient concepts from the retrieved queries from an online knowledge repository,determining the central concepts of the original query from the extracted salient concepts, said central concepts being determined from a relationship graph of salient concepts for said queries retrieved from said historical query log, andusing said determined central concepts to construct a revised query; submitting the revised query to the search engine; and providing search results for the revised query to the user. |
