Method and device for searching people with same name

Abstract

The embodiment of the invention discloses a method and a device for searching people with the same name. The method comprises the following steps of: retrieving an input target person name, and thus obtaining relevant documents which contain the target person name; respectively extracting people relation characteristic information in each relevant document, counting the people relation characteristic information in each relevant document, establishing a people relation graph, and calculating the relation strength of the target person name and other people names in the people relation graph; establishing a people relation characteristic vector for each relevant document according to person name which is contained in each relevant document and the relation strength; and according to the people relation characteristic vector, clustering all relevant documents to obtain a people relation clustering result, so a searching result is at least given according to the people relation clustering result. By the invention, the clustering accuracy of the people with the same name can be improved, so the quality of the searching result is improved, and the performance of a searching engine is improved.

Description

technical field [0001] The present invention generally relates to the technical field of search, in particular to a search method and system for people with the same name. Background technique [0002] With the development of the network, there are more and more information with the same name on the network, how to distinguish these people with the same name is a problem that needs to be paid attention to in the process of information search. For example, when a user enters a person's name in a search engine, the returned results contain information about many people with the same name, and the user needs to click and browse one by one to determine which information is really needed, which is obviously time-consuming and laborious. Of course, users can also narrow down the search scope by artificially adding query words and other methods. However, manually adding query words is equivalent to filtering the query results, which reduces the coverage of the query. Therefore, t...

AI Technical Summary

Problems solved by technology

[0005] In short, in the prior art, when relying on the similarity between document topics to cluster webpages containing people with the same name, there may be the problem of inaccurate clustering, and then when providing search results based on the clustering results, there will also be Makes the quality of search results low, which in turn limits the improvement of search engine performance

Images