Individual Doctor Recommendation in Large Networks by Constrained Optimization

In this paper, the authors try to systematically investigate the problem of individual doctor recommendation and propose a novel method to enable patients to access such intelligent medical service. In their method, the authors first mine doctor-patient ties/relationships via Time-constraint Probability Factor Graph model TPFG from a medical social network. Next, they design a constraint-based optimization framework to efficiently improve the accuracy for doctor-patient relationship mining. Last, they propose a novel Individual Doctor Recommendation Model, namely IDR-Model, to compute doctor recommendation success rate based on weighted average method. The authors conduct experiments to verify the method on a real medical data set. Experimental results show that they obtain better accuracy of mining doctor-patient relationship from the network, and doctor recommendation results of IDR-Model are reasonable and satisfactory.