Medical consultation guide generation method based on topic models and ILP

Abstract

The invention discloses a medical consultation guide generation method based on topic models and ILP. The steps are as follows: (1) disease knowledge and medical consultation experience data issued byusers are acquired, and a domain knowledge dictionary is obtained from the Sougou word library, a traditional Chinese medicine subject dictionary and the Baidu encyclopedia; (2) after duplicates areremoved, the medical consultation experience data are merged into a library; (3) the disease knowledge is utilized for creating a guide template for diseases, and the guide template contains nine major subject forums; (4) entity identification is carried out on medical consultation experiences; (5) joint subject modeling is carried out on the guide template and the medical consultation experiencescontaining entities, and topic-word distribution and topic-entity distribution are generated; (6) the topic-word distribution is used for choosing a plurality of medical consultation experiences foreach subject forum; (7) the medical consultation experiences in step (6) are divided into sentences; (8) the topic-word distribution and the optic-entity distribution are used for carrying out ILP optimization solution, and thereby medical consultation guide abstracts of the nine major subject forums for each disease are obtained.

Description

technical field [0001] The invention relates to the fields of entity recognition, topic model, and automatic text summarization, in particular to a method for generating medical treatment strategies based on topic model and ILP. Background technique [0002] The Internet + medical model is rapidly changing the way the public visits doctors to find medicines. More and more users will publish experience sharing content based on real medical experience on doctor-patient communication platforms and communities, and share disease knowledge and rehabilitation experience with other users. However, a large amount of heterogeneous and heterogeneous experience sharing data has not been effectively sorted out and analyzed, and knowledge mining has not yet been carried out, making it difficult for users to quickly obtain the desired medical knowledge and experience in seeing a doctor. In view of the above, this patent focuses on entity recognition of medical experience by using conditio...

AI Technical Summary

Problems solved by technology

[0002] The Internet + medical model is rapidly changing the way the public visits doctors to find medicines. More and more users will publish experience sharing content based on real medical experience on doctor-patient communication platforms and communities, and share disease knowledge and rehabilitation experience with other users. Discussion and exchange, but a large amount of heterogeneous and heterogeneous experience sharing data has not yet been effectively sorted out and analyzed, making it difficult for users to quickly obtain the desired medical knowledge and experience in seeing a doctor

Images