Entity linkage algorithm based on graph model

Abstract

The present invention discloses an entity linkage algorithm based on a graph model. The entity linkage algorithm based on a graph model is characterized by comprising: forming a candidate entity by using the Wikipedia knowledge base; constructing a semantic feature between the entities by using LDA; constructing relationships between entities based on linkage structures of Wikipedia to form a graph model; and integrating related semantic features into the graph model; and ranking the entities by using the PageRank algorithm to obtain an entity linkage result, which specifically comprises steps of calculation and integration of a naming dictionary, a candidate entity set, related features, construction of the graph model, and ranking of candidate entities. Compared with the prior art, the entity linkage algorithm based on a graph model has the advantages of being good in entity feature integration and high in reliability of the entity linkage result; data is downloaded by using Wikipedia, so that no additional costs are needed, and especially data sets do not need to be noted manually; and the method is simple, convenient in usage, and saves time and efforts.

Description

technical field [0001] The invention relates to the technical field of information base text processing, in particular to an entity linking algorithm based on a graph model. Background technique [0002] The research objects of the entity linking task are entity nouns that include three types of tasks, institutions, and places. The research goal is: given a query that contains the target entity and the background documents that support the query word, combine the secondary target entity with the existing knowledge The entities with mutual referential relationship in the knowledge base are correctly connected. If there is no entity node connected with the query entity in the knowledge base, it is called a non-KB entity, and such non-KB entities are clustered. Add the entities required by the universal query to the knowledge expansion, and expand and maintain the knowledge base. Therefore, on the one hand, the entity linking task can accurately feed back the user's query resu...

AI Technical Summary

Problems solved by technology

The disadvantage is that this method requires manual labeling of the data set, which requires a lot of manpower and time, and does not mine some semantic information describing the target entity in the article, but treats all entities appearing in the article equally. The role of semantic information of the target entity is ignored

The advantage of the unsupervised learning method is that it does not need to label data, which saves a lot of manpower and time. The disadvantage is that the features are not easy to integrate

Images