Knowledge graph fusion method based on short text similarity calculation

Abstract

The invention discloses a knowledge graph fusion method based on short text similarity calculation, and belongs to the crossing field of education and computer technologies. Knowledge graph fusion mainly exists in a graph fusion form in practical application, and in existing work. Most of knowledge graph fusion focuses on a topological structure in a knowledge graph and ignores semantic information carried in entity name literal meanings. In order to solve the problems, the invention provides a knowledge graph fusion method based on short text similarity calculation. The crowdsourcing thoughtis combined, individual knowledge sub-graphs are obtained, and after the sub-graphs are obtained, the knowledge sub-graphs are preprocessed mainly through sub-graph noise reduction and similarity calculation and then knowledge graph fusion is carried out. The invention provides a specific knowledge graph fusion method, and compared with the traditional similar knowledge graph fusion method, the method provided by the invention has higher correctness and effectiveness, and finally, an effective graph set can be obtained.

Description

technical field [0001] The invention relates to a knowledge map fusion method based on short text similarity calculation, which belongs to the intersecting field of education and computer technology. Background technique [0002] As the data support of the Semantic Web, knowledge graphs are widely used in fields such as semantic search, in-depth question answering, and online education. In terms of education, the current mainstream is still to use a test paper to judge a learner's mastery of the course and the learning status of the quarter, but it is impossible for learners to have personalized and diversified learning methods, and it is impossible for students to be more knowledgeable. Actively participate in the process of learning knowledge. [0003] The knowledge graph is essentially a semantic network, which clearly expresses the entities and their relationships in the physical world. At present, a large number of knowledge graph-related products have emerged, among ...

AI Technical Summary

Problems solved by technology

In terms of education, the current mainstream is still to use a test paper to judge a learner's mastery of the course and the learning status of the quarter, but it is impossible for learners to have personalized and diversified learning methods, and it is impossible for students to be more knowledgeable. Actively participate in the process of learning knowledge

Images