A medical specialty outpatient clinics recommendation system based on text mining

Many prospective medical patients have difficulty determining which type of outpatient specialist to consult for their complaints, and their resulting inquiries impose an additional administration cost for hospitals. The symptoms data of outpatient treatment in different specialties are collected from various hospitals to establish a database fronted by a chatbot-based interface to develop a medical specialty outpatient clinic recommendation system. The proposed system integrates speech recognition, the Jieba word segmentation algorithm and the conditional random field algorithm to retrieve keywords during the dialogue process. The C4.5 decision tree model is used to provide clinical department referrals for the symptoms reported by patients. Through continuous revision, the system gradually reduces the error rate of outpatient recommendations, thus reducing patients' waiting times and the workload of frontline hospital staff. Eleven subjects were invited to use this system, and seven of them felt that this system could help them.