Font Size: a A A

Research On Automatic Acquisition Of TCM Symptom Terms

Posted on:2019-01-04Degree:MasterType:Thesis
Country:ChinaCandidate:Q WangFull Text:PDF
GTID:2404330566974019Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
Traditional Chinese Medicine(TCM)symptom information is the basic data source of clinical research.In order to promote the informatization process of TCM and dig clinical knowledge of TCM further more,a large number of terminology of clinical symptoms are needed as support.The automatic acquisition of terms is a difficult part of natural language processing.Due to the particularity of Chinese medical documents,it is a research challenge to automatically acquire TCM terms from Chinese medical documents.On the basis of analyzing the characteristics of TCM symptom terms,this paper uses natural language processing technology to automatically obtain TCM symptom terms from the clinical medical records of TCM.In view of the lack of existing clinical terminology resources in TCM,and there are some technical difficulties to automatically acquire terminology.This dissertation draws on the methodology of named entity recognition and combines the characteristics of TCM clinical medical records.The main research contents are as follows:1.Constructing the corpus suitable for the acquisition of symptom terms with reference to the construction method and annotation specification of the existing TCM corpus;2.By analyzing the constitutional patterns of symptom terms in TCM clinical records,a method of using term formation patterns to automatically obtain TCM symptom terms was proposed.3.Aiming at the problem of low recall rate based on pattern auto-acquisition,this paper proposes a method based on a combination of generalization patterns and statistical features to automatically obtain symptoms terminology from the clinical records of TCM.The experimental results show that the method has an accuracy rate of 81.1%,a recall rate of 82.2% and an F value of 81.6%.4.After obtaining a large number of clinical symptom terms,combined with the terms of TCM clinical symptoms nomenclature and TCM symptomology,we used the ontology construction method to classify the four diagnostic methods Concept,build the TCM symptom ontology knowledge representation model.
Keywords/Search Tags:TCM symptom terms, part-of-speech pattern, field characteristic words, symptom ontology
PDF Full Text Request
Related items