Text data processing method, apparatus, electronic device and computer readable medium

Abstract

The present disclosure relates to a text data processing method, an apparatus, an electronic device and a computer readable medium. The invention relates to the field of medical information processing. The method comprises the following steps: preprocessing text data to generate a plurality of text units, wherein the data in the paper is unstructured data; Processing the text data according to a predetermined dictionary and set to extract entity words, type words, and expressions; Generating a set of text sequences according to the plurality of text units and the entity words, type words, andexpressions; And performing pattern matching and disassembly based on the set of text sequences to generate a structured data result. The disclosed text data processing method, device, electronic device and computer-readable medium can accurately and efficiently convert unstructured text data into structured data, thereby improving medical data processing efficiency.

Description

technical field [0001] The present disclosure relates to the field of computer information processing, in particular, to a text data processing method, device, electronic equipment and computer readable medium. Background technique [0002] Electronic medical records record a large amount of real and rich clinical data, which is a summary of clinicians' long-term practice and experience, and can be used to support clinical decision-making, epidemiological statistics, clinical research and drug development, etc. However, there are a large number of unstructured texts based on natural language in electronic medical records. If this part of data can be used and understood by computers, it needs to extract structured information based on clinical natural language processing. On the other hand, different hospitals, different disciplines, and different doctors record medical records and clinical terms in different ways of expression and writing habits, which brings challenges to t...

AI Technical Summary

Problems solved by technology

However, the existing rule-based methods are mainly based on manual experience summarization, and do not have the advantages of combining machine mining; relationship classification based on machine learning relies on more manually labeled data, and the difference in data distribution makes it difficult to quickly adapt to cross-institution data; finally, the existing rule / machine learning-based relationship recognition stays on the basis of binary relationship classification, and there are many problems in the actual application process, such as relationship conflicts, relationship hierarchical combination, etc.

Images