| 摘要 |
Disclosed are a search recommending method and device based on search terms. The method comprises: constructing an anonymous behavior network topology according to an anonymous search log; and selecting recommendation results related to search terms from the anonymous behavior network topology in response to an event of inputting the search terms by a user. According to the technical solution provided by the present invention, extension is performed on the basis of user behaviors, minority dimensions that the users are most concerned with are mined deeply, the defect that the recommendation results are excessively generalized in the prior art is eliminated, the recommendation results are divided through clustering, each class is accurately described by using knowledge graphs, and valuable information is provided, so that the selected recommendation results better conform to search habits and search demands of the users. |
