基于多标记学习的中医药方功能预测研究OACSTPCD

Different from the western medicine,the Traditional Chinese Medicine(TCM)has created an independent medi-cine system.During the past thousands years,TCM has accumulated a mass of mature prescriptions.Then,how to extract some po-tentially valuable knowledge from these TCM prescriptions has attracted many researchers in machine learning and data mining fields.This study considers to construct the mapping relationship between TCM prescription and its efficacies and to develop approxi-mately accurate TCM prescription efficacy prediction model by using multi-label learning techniques.The experiments are conduct-ed on multiple different multi-label learning algorithms and more than 1000 mature TCM prescriptions.The results show that the multi-label learning technique is capable of providing approximately accurate prediction for TCM prescriptions'efficacies.