Development of models for classification of action between Heat-Clearing herbs and Blood-Activating Stasis-Resolving herbs based on the theory of Chinese herbal property
Action of Chinese herbs is the high recapitulation for therapeutic and health-preserving effects under the guidance of Chinese herbs theory.Traditional Chinese medicine herbal properties(TCM-HPs)are the high generalization and summary of the Chinese herbal actions.In this research,we constructed the classification models of Heat-Clearing herbs(HCHs)and Blood-Activating Stasis-Resolving herbs(BASRHs)to predict the herbal actions based on the theory of Chinese herbal property.Our team also analyzed distribution patterns of TCM-HPs of HCHs and BASRHs to detect the feature properties of these two types of herbs.In addition,we used machine learning(ML)and deep learning(DL)methods to examine whether the action of Heat-Clearing and Blood-Activating Stasis-Resolving could be predicted correctly based on the theory of Chinese herbal property.Chinese herbal properties information on 107 herbs including 74 HCHs and 33 BASRHs were collected,which were split into 80 training samples and 27 test samples using the Kennard-Stone(K-S)algorithm.Four types of methods correctly predicted 48.15–85.19%of 27 test samples including 13 HCHs and 14 BASRHs.The method,Deep Belief Network(DBN)gave the best results,it correctly predicted 100.00%of the 13 HCHs and 71.41%of 14 BASRHs.Our study suggest that we could adopt machine learning and deep learning methods to predict the actions of Heat-Clearing and Blood-Activating Stasis-Resolving based on the theory of Chinese herbal property.This research used the machine learning and deep learning methods to explain the connotation of Chinese herbal properties-actions(CHPs-As)scientifically,and to improve our understanding of the relationship of Chinese herbal properties-actions(CHPs-As).After that we will constantly enrich and develop the Chinese herbal theory.
Heat-Clearing herbs Blood-Activating Stasis-Resolving herbs Machine learning Deep learning Chinese herbal property Traditional Chinese medicine
Zhao Chen Yanfeng Cao Shuaibing He Yanjiang Qiao
School of Chinese Materia Medica,Beijing University of Chinese Medicine
国内会议
合肥
英文
1-28
2017-06-10（万方平台首次上网日期，不代表论文的发表时间）