Post-structured knowledge discovery method and device for electronic medical record

Abstract

The invention relates to a post-structured knowledge discovery method and device for an electronic medical record.The method includes the steps that original electronic medical record text corpus data is written into a text corpus, fact element corpora in the text corpus are read through a fact element logic reader-writer and written into a fact element base, event element corpora in the fact element base are read through an event element logic reader-writer and written into an event element base, and fact structure corpora in the event element base are read through a fact structure logic reader-writer and written into a fact structure base, wherein 'ultrasonic data and sensitive data point-to-group data lock rule between sensitivity numbers and hyper numbers' is deployed on a rule base of the even element logic reader-writer.Based on number theory research, a unique data lock structure in the hypersensitivity law is utilized for converting simple and mechanical language-to-language robot learning ability into fact-to-language knowledge discovery ability, automatic management of the knowledge discovery process is achieved with the help of the ability, and therefore the scientific research management level of medical education can be substantially improved.

Description

technical field [0001] The invention relates to the field of medical education and scientific research management, in particular to a method and device for discovering structured knowledge after electronic medical records. Background technique [0002] With the popularization of computer applications, the management of medical knowledge includes machine summarization of known knowledge and machine extraction of unknown knowledge, which has become a key technical means for the development and progress of medicine. [0003] Electronic medical records are created, stored and used electronically by medical institutions, focusing on data integration systems for clinical diagnosis and treatment of outpatients and inpatients and guidance for intervention information. It is a complete and detailed clinical information resource generated and recorded by individual residents during their previous medical visits in medical institutions. [0004] With the development of the Internet of...

AI Technical Summary

Problems solved by technology

[0007] However, various natural language processing technologies strive to present the machine understanding ability of language itself, only focus on the machine's extraction and recognition of grammar and semantics, and have no awareness of the factual environment (context) that language depends on. The current natural language Processing technology focuses on and emphasizes "extraction of language expression rules" while ignoring "extraction of factual elements to be expressed in language". This is a machine learning method for language for language, not a knowledge discovery method for language for facts

Images