Entity relationship automatic labeling method applied to medical text

Abstract

The invention discloses an entity relationship automatic labeling method applied to a medical text. The entity relationship automatic labeling method comprises the following steps: S1, constructing amedical term dictionary and a priori knowledge base, S2, performing statistical co-occurrence on the target medical text according to the medical term dictionary and generating a basic corpus, S3, pre-annotating the basic corpus by adopting a priori knowledge base to obtain a pre-annotated corpus, S4, performing entity correction on the pre-annotated corpus to obtain an entity-corrected corpus, and S5, filtering the corpus subjected to entity correction through relationship marker words to finish automatic labeling of the entity relationship of the medical text. According to the invention, anautomatic labeling form is adopted, the labor input of researchers in relation extraction is reduced, the labeling speed is high, meanwhile, the method does not need to depend on labeling of experts,the problem that medical texts are difficult to label is solved, the labeled texts are further filtered through relation marker words, and the labeling precision can be obviously improved.

Description

technical field [0001] The invention relates to the field of natural language processing, in particular to an automatic entity relationship labeling method applied to medical texts. Background technique [0002] With the continuous development of the field of precision medicine, there is an increasing trend of papers related to topics such as disease genes every year. Papers are the main carrier of relational knowledge in precision medicine, and the automatic extraction of structured information from them is the main factor that promotes the development of precision medicine. [0003] Linking human diseases with the genes, drugs, etc. involved is the core of precision medicine. These connections can be made through a variety of different types of studies, including classic lineage genetics studies of Mendelian and complex diseases, genome-wide association studies (GWAS), online Mendelian inheritance in humans, somatic mutation frequencies, transcriptomics and proteomics sc...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the above-mentioned deficiencies in the prior art, the present invention provides a method for automatic labeling of entity relations applied to medical texts, which solves the problem of low accuracy of existing methods for automatic extraction of entity relations in medical texts

Images