Open AccessResearch on Similar Medical Records Recommend Model based on Natural Language Processing
Author(s) -
Yong Shuai,
Chunjiang Zhou,
Qiaoyue Pang,
Chao Yuan,
Zheng Xie
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1550/3/032105
Subject(s) - computer science , medical record , credibility , similarity (geometry) , complaint , information retrieval , data mining , medical history , natural language , natural language processing , artificial intelligence , medicine , radiology , political science , law , image (mathematics)
Aiming at the problem of low efficiency and poor credibility in medical record text data mining, this paper proposes a similar medical record recommend model based on natural language processing. Firstly, extract the samples by structured data in the medical record data and narrow down the extraction of similar medical records, then use the text keyword extraction algorithm to obtain the keyword text data about chief complaint, medical history and physical examination in the medical record text, use these data as input data, get the similarity between the input data and the patient data by text similarity algorithm, and use collaborative filtering algorithm to calculate the most similar medical records and obtain the most similar medical records diagnosis conclusions, take the chief complaint, medical history, physical examination data and diagnosis conclusions as input data to calculate the text similarity and collaborative filtering values and obtain the final recommend treatment plan. The case analysis proves that the model proposed in this paper has high credibility and practicality.