Knowledge graph expansion method, electronic equipment and storage medium

Abstract

The embodiment of the invention relates to the technical field of knowledge maps, and discloses a knowledge graph expansion method, electronic equipment and a storage medium. The method comprises thefollowing steps: obtaining a keyword, finding an ontology where the keyword is located in a preset database, and locating the keyword according to the ontology where the keyword is located and the knowledge map. Obtaining a first class of statements and a second class of statements in to-be-processed text data, marking a first triple corresponding to the first class of statements according to theknowledge graph, and training by utilizing the first class of statements marked with the first triple to learn an association relationship between the first class of statements and the first triple, obtaining a relationship identification model, identifying the second class of statements by using the relationship identification model, determining a second triple corresponding to the second class of statements, and finally adding the second triple corresponding to the second class of statements to the knowledge graph. That is to say, triples in a certain field can be automatically extracted through keywords in the field and added into the knowledge graph, and therefore the knowledge graph is expanded.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of knowledge graphs, and in particular, to a method for expanding knowledge graphs, electronic devices, and storage media. Background technique [0002] As the most convenient information acquisition platform today, the Internet has become increasingly urgent for users to obtain effective information through the Internet. How to obtain effective value information from massive data has become a difficult problem, and knowledge graphs have emerged as the times require. [0003] Knowledge graph (knowledge graph, KG) is essentially a semantic network, a graph-based data structure composed of nodes and edges. In the knowledge graph, each node represents an entity, and each edge represents the relationship between entities. Among them, entity refers to something that is distinguishable and exists independently, such as a certain person, a certain city, a certain kind of plant, a certain kind...

AI Technical Summary

Problems solved by technology

However, the coverage of knowledge graphs for entities in a certain field is often incomplete, and the number of entities recorded in knowledge graphs is often smaller than the number of entities that actually exist in a certain field. Therefore, it is necessary to continuously expand knowledge graphs

In the existing technology, the relationship set and the entity pair set are usually defined separately, that is, the processes of schema construction, entity recognition, and relationship extraction are processed separately, and the connection between two pairs must be repeatedly executed every time a triple is extracted. , cannot automatically extract triples and add

Images