Method and device for extracting relationships among entities in traditional Chinese medicine documents

Abstract

The invention provides a method and a device for extracting a relationship between entities in traditional Chinese medicine documents, which can improve the accuracy of extracting a relationship typebetween entities. The method comprises the following steps: for a to-be-processed traditional Chinese medicine literature, obtaining an entity type and an entity-to-entity relationship type marked forpartial contents of the to-be-processed traditional Chinese medicine literature; training a named entity recognition model according to the marked entity type; utilizing the trained named entity recognition model to carry out named entity recognition on the to-be-processed traditional Chinese medicine literature, and obtaining a candidate entity pair and a feature table with a relation accordingto a named entity recognition result; according to the obtained candidate entity pairs with the relations and a feature table, using a factor graph model to carry out statistical reasoning on graph probabilities, learning entity relation features globally, and obtaining the probability of the relations between entities; and determining the type of the relationship between the entities according tothe obtained probability of the relationship between the entities in combination with a dependency analysis method for extracting the fact triple and the marked relationship type between the entities. The invention relates to the field of knowledge engineering.

Description

technical field [0001] The invention relates to the field of knowledge engineering, in particular to a method and device for extracting relationships between entities of Chinese medicine documents. Background technique [0002] There are many ancient books and documents handed down in the field of Chinese medicine in my country, which are the basic basis for learning Chinese medicine. However, most of these documents are written in ancient Chinese, and most of them are unstructured texts, which are very time-consuming to use. If each entity and the entity relationship among them can be extracted from TCM literature, the relationship between the extracted entities can be used to effectively carry out information retrieval, knowledge mining and so on. [0003] The entity relationship extraction methods in the prior art are difficult to accurately extract the relationship between entities from unstructured text. Contents of the invention [0004] The technical problem to be...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to provide a method and device for extracting the relationship between entities in traditional Chinese medicine literature, so as to solve the problem existing in the prior art that it is difficult to accurately extract the relationship between entities from unstructured texts

Images