Entity set expansion method and apparatus

Abstract

Embodiments of the invention provide an entity set expansion method and apparatus. The method comprises the steps of extracting candidate entities from a target knowledge graph to form a candidate entity set according to a pre-determined seed entity set; determining meta-paths between seed entities from a heterogeneous information network corresponding to the target knowledge graph, wherein each meta-path is a connection path, consisting of an entity type and a relational type, between two node types in the heterogeneous information network, and the two node types are node types corresponding to different seed entities; determining a first importance degree of each meta-path according to a quantity of seed entity pairs connected by each meta-path; according to the first importance degree of each meta-path, determining a second importance degree of each candidate entity in the candidate entity set; and determining the candidate entity with the second importance degree meeting a first preset condition in the candidate entity set as a to-be-expanded entity, and adding the to-be-expanded entity to the seed entity set. By applying the method and the apparatus, effective entity set expansion can be carried out.

Description

technical field [0001] The invention relates to the technical field of entity set extension, in particular to an entity set extension method and device. Background technique [0002] Entity set expansion means that several entity seeds with a specific semantic type (also called a specific common feature) are known, and more entities of the specific semantic type are obtained according to certain rules. For example, given the entity seed set {Beijing, Washington, Moscow} whose specific semantic type is the capital of a country, it is required to find out more capitals of the country, such as finding {Seoul, Tokyo, Kuala Lumpur,...}. At present, entity set expansion has been widely used, for example, dictionary expansion and query suggestion expansion. [0003] The most common method of entity set expansion is to select a data source, process the data source according to certain rules, and determine other entities with the same semantic type as the seed entity as the expansio...

AI Technical Summary

Problems solved by technology

However, due to the limited amount of data contained in a single text and web page, the effectiveness of entity set expansion is not ideal, and it cannot meet the increasing demand for entity set expansion.

Images