Knowledge graph generating method and device

Abstract

The invention provides a knowledge graph generating method and device and belongs to the technical field of computers. The method includes: acquiring meta-event information corresponding to each data source from at least two data sources; extracting preset information corresponding to the meta-event information from the meta-event information corresponding to each data source, wherein the preset information includes an entity and attribute thereof; performing association and fusion on the preset information to generate a knowledge graph. By the knowledge graph generating method and device, efficiency of a user in acquiring information is improved.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a method and device for generating a knowledge graph. Background technique [0002] With the rapid development of the Internet, people have more and more diverse ways to obtain information and knowledge, but massive data are scattered in every corner of the Internet, which brings great obstacles to users to obtain the information they really need. [0003] At present, the fields of news and social media event detection focus more on a single event or a simple relationship between people. For general events and events, the relationship between people and events often requires manual editing and sorting (for example, thematic sorting of news portals). In most cases, users The results returned when searching for relevant events or news are still independent web pages with high redundancy and low precision. There are big problems in recall rate and precision rate. Users oft...

AI Technical Summary

Problems solved by technology

[0002] With the rapid development of the Internet, people have more and more diverse ways to obtain information and knowledge, but massive data are scattered in every corner of the Internet, which has brought great obstacles to users in obtaining the information they really need.

[0003] At present, the fields of news and social media event detection focus more on a single event or a simple relationship between people. For general events and events, the relationship between people and events often requires manual editing and sorting (such as thematic sorting of news portals). In most cases, users The results returned when searching for related events or news are still independent web pages with high redundancy and low precision, which have big problems in recall and precision, and users often need to spend more time browsing to find what they want information and knowledge

[0004] Therefore, using the existing data processing and information retrieval methods, the efficiency of users to obtain information is not high

Images