A vector space model for automatic indexing
Communications of the ACM
Modern Information Retrieval
Optimizing search engines using clickthrough data
Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining
Training linear SVMs in linear time
Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining
Voting for candidates: adapting data fusion techniques for an expert search task
CIKM '06 Proceedings of the 15th ACM international conference on Information and knowledge management
ArnetMiner: extraction and mining of academic social networks
Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining
Mining advisor-advisee relationships from research publication networks
Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining
Re-examination of interestingness measures in pattern mining: a unified framework
Data Mining and Knowledge Discovery
Factor graphs and the sum-product algorithm
IEEE Transactions on Information Theory
Reduced Support Vector Machines: A Statistical Theory
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
It is difficult for patients to find the most appropriate doctor/physician to diagnose. In most cases, just considering Authority Degrees of Candidate Doctors(AD-CDs) cannot satisfy this need due to some objective preferences such as economic affordability of a patient, commuting distance for visiting doctors and so on. In this paper, we try to systematically investigate the problem and propose a novel method to enable patients access such intelligent medical service like this. In the method, we first mine patient-doctor relationships via Time-constraint Probability Factor Graph mode(TPFG) from a medical social network, and then extract four essential features for AD-CDs that would be subsequently sorted via Ranking SVM. At last, combining AD-CDs and patients' preferences together, we propose a novel Individual Doctor Recommendation Model, namely IDR-Model, to compute doctor recommendation success rate based on weighted average method. We conduct experiments to verify the method on a real medical data set. Experimental results show that we obtain the better accuracies of mining patient-doctor relationship from the network, AD-CDs ranking is also better than the traditional Reduced SVM method, and doctor recommendation results of IDR-Model is reasonable and satisfactory.