Entity linking method in a knowledge base question answering system

Abstract

The invention provides an entity linking method in a knowledge base question answering system, which comprises the following steps: obtaining a subject term set from a question sentence; searching inthe knowledge base according to the obtained set of subject terms to obtain a preliminary set of candidate entities; For each entity in the preliminary candidate entity set, extracting a correspondingfeature from the entity, the question and the knowledge base; and obtaining a score of the entity according to the extracted feature of each entity in the preliminary candidate entity set, and obtaining a candidate entity set according to the score. The invention improves the accuracy and efficiency of the entity link.

Description

technical field [0001] The invention relates to the technical field of the Internet, in particular to an entity linking technology in a question answering system of a knowledge base. Background technique [0002] The knowledge base question answering system is a widely used question answering system at present. It analyzes and infers the natural language questions input by the user, and combines the content in the knowledge base to search for the correct answer, and then directly gives the answer according to the time and space scenes used. Question answers and related information can well meet people's requirements for accurate information. As a key technology in knowledge base question answering system, entity linking has become one of the important factors affecting the effect of knowledge base question answering system. Entities refer to things that can be distinguished from each other and exist independently, while entity linking refers to linking the subject words in ...

AI Technical Summary

Problems solved by technology

[0004] However, when extracting subject words, in order to ensure the recall rate of correct entities, existing methods generally extract a large number of incorrect candidate subject words, which not only brings a lot of noise to the subsequent steps, but also seriously affects entity links. method performance

In addition, when screening the candidate entities obtained from the knowledge base search, only some simple symbol-level methods and rule items are used. In a knowledge base with a large number of entities, similar entities (for example, entities with the same name, namely name The same but different attributes; the same attributes but different names; most of the attributes are the same and the same names, etc.) to effectively distinguish, which has a greater impact on the subsequent operations of the knowledge base question answering system (such as obtaining the correct semantic relationship)

Images