Tail entity linking method and device, server and storage medium

Abstract

The embodiment of the invention provides a tail entity linking method and device, a server and a storage medium, and the method comprises the steps: obtaining a knowledge graph which comprises a headentity, a to-be-linked tail entity and an entity relation describing the head entity and the to-be-linked tail entity, and judging whether the to-be-linked tail entity is a single entity or not through a single entity classification model; if the to-be-linked tail entity is a single entity, determining a candidate linked entity set corresponding to the single entity; wherein the candidate linked entity set comprises N candidate linked entities, and N is a positive integer; calculating the similarity between the single entity and each candidate linked entity in the N candidate linked entities through a similarity model; and if the similarity between the single entity and the target candidate linked entity is greater than a first threshold, linking the target candidate linked entity with thesingle entity. According to the embodiment of the invention, the accuracy of the tail entity link can be improved.

Description

technical field [0001] The present application relates to the technical field of machine learning, in particular to a tail entity linking method, device, server and storage medium. Background technique [0002] With the development of technology and the popularization of the Internet, we have entered the era of big data. Although there are a large number of data resources in the network, these data are heterogeneous and diverse, and the organizational structure is loose, so they cannot be effectively utilized. In order to obtain information and form knowledge, people have proposed knowledge graphs. Knowledge graph (knowledge graph, KG) has laid a foundation for the organization and intelligence of network data with its powerful semantic processing ability and open organization ability. At present, knowledge graph, as one of the key technologies in the development and application of artificial intelligence technology, has been widely used in intelligent search, intelligent q...

AI Technical Summary

Problems solved by technology

[0004] In the current process of entity linking, manually labeled features are usually input into the traditional machine learning model to calculate the similarity score. These features are mostly used to count the number of co-occurrence windows, resulting in less information that the model can obtain and larger information entropy. Links are less accurate

Images