ERNIE based intelligent triage system

Article type: Research Article

Authors: Wang, Chuantao^{a; b} | Feng, Fan^{a; b; *}

Affiliations: [a] School of Mechanical-Electronic and Vehicle Engineering, Beijing University of Civil Engineering and Architecture, Beijing, China | [b] Beijing Engineering Research Center of Monitoring for Construction Safety, Beijing, China

Correspondence: [*] Corresponding author. Fan Feng, Tel.: +86 18935469496; E-mail: fengnaf@126.com.

Abstract:  With the development of Internet+medicine, online medical treatment has gradually become the new development direction of medical industry. Many hospitals provide online registration services to the public, and due to the lack of professional medical knowledge of patients, the problem of wrong registration often occurs. How to use deep learning technology to provide professional help to patients and reduce the waste of medical resources has become an urgent problem. To address the above problems, this paper proposes an ERNIE-based text classification model for intelligent triage. The model consists of two parts, ERNIE and BiGRU. The pre-training model ERNIE is used to extract the feature representation of the text, and then input to the BiGRU neural network to get the text classification results. Compared with different models on 2 datasets, the experimental results show that the model proposed in this paper has better accuracy and recall than other models.

Keywords: Text classification, ERNIE, deep learning, intelligent triage

DOI: 10.3233/JIFS-212140

Journal: Journal of Intelligent & Fuzzy Systems, vol. 43, no. 4, pp. 5013-5022, 2022