Doctor Recommendation Based on Patient Syndrome Using Convolutional Neural Network

Recommendation systems in the online medical sector assist patients in finding appropriate doctors. This paper aimed to solve the complication in doctors' recommendations, concerning that people often struggle to see sure doctors according to their medical needs. Currently, most existing systems create doctors' recommendations through explicit or implicit feedback mechanisms. This doctor recommendation model does not depend on user feedback; instead, candidate doctors are generated for guidance solely from the user's current medical conditions. A prognosis is predicted for a specific syndrome via CNN. By applying discrete rules, the system identifies and fetches the most relevant specialists according to the prediction and provides necessary information. The performance evaluation results of the proposed method are high and satisfactory.