基于改进关联规则算法的中医处方规律挖掘研究
|
摘要
目的:通过增加中医处方药物的剂量特征改进关联规则算法以探寻处方规律。方法:将处方药物的平均剂量以及配伍减毒增效关系纳入关联规则算法改进模型中,分别构建不同加权支持度,用Python实现改进后的算法,并对结果进行可视化,最后在中医临床专家的指导下论证模型的有效性。结果:根据2种加权方法设计了2种改进挖掘模型,在相同支持度下,2种改进模型均获得了比传统模型更多的药物组合,其中配伍减毒增效加权模型效果更为显著。结论:使用本文提出的改进挖掘模型可以挖掘到更多频繁项集,减少用药规律的遗漏,对中医处方规律研究有较强实用价值。
Objective: By improving the existing association rules algorithm through increasing the dose characteristics of the prescription drugs of traditional Chinese medicine(TCM) to find out the rules of the prescription better. Methods:First, the dose-weighted support was constructed based on the relationship between the average dose of prescription drugs and the drug-attenuated synergistic effect. Then, based on different support degrees, the improved association rule models were implemented using Python and visualize the results. Finally, the validity of the model was demonstrated under the guidance of TCM clinical experts. Results: Two kinds of improved mining models based on association rules were constructed respectively based on two kinds of weighting methods. Under the same support degree, two kinds of improved models obtained more drug combinations than the traditional model, and the second new model is more significant. Conclusion: By using the improved mining model proposed in this paper, more frequent itemsets can be mined and fewer medication rules will be missed, which has a strong practical value for the study of the rule of TCM prescription.
Objective: By improving the existing association rules algorithm through increasing the dose characteristics of the prescription drugs of traditional Chinese medicine(TCM) to find out the rules of the prescription better. Methods:First, the dose-weighted support was constructed based on the relationship between the average dose of prescription drugs and the drug-attenuated synergistic effect. Then, based on different support degrees, the improved association rule models were implemented using Python and visualize the results. Finally, the validity of the model was demonstrated under the guidance of TCM clinical experts. Results: Two kinds of improved mining models based on association rules were constructed respectively based on two kinds of weighting methods. Under the same support degree, two kinds of improved models obtained more drug combinations than the traditional model, and the second new model is more significant. Conclusion: By using the improved mining model proposed in this paper, more frequent itemsets can be mined and fewer medication rules will be missed, which has a strong practical value for the study of the rule of TCM prescription.
引文
1唐仕欢,杨洪军.中医组方用药规律研究进展述评.中国实验方剂学杂志, 2013(5):359-363.
2 田瑾.基于复杂网络及关联规则的失眠用药中医临床数据挖掘研究.北京:北京中医药大学硕士研究生学位论文, 2015.
3 仝小林.论辨症、辨病、审因与辨证论治在临床中的应用.中医杂志,2013(2):93-95.
4 谢宁,郑洪新,赵文静.中医学基础.北京:中国中医药出版社, 2011(2):23-26.
5 汤尔群,黄玉燕,桑希生.《黄帝内经》组方理论探讨.安徽中医学院学报, 2012(1):1-3.
6 Brin S, Motwani R, Silverstein C. Beyond market basket:Generalizing association rules to correlations. Management of Data(SIGMOD97),1997:265-276.
7 韩家炜, Kamber M.数据挖掘概念与技术.北京:机械工业出版社,2012(7):55-61.
8 Han J, Pei J, Yin Y. Mining frequent patterns without candidate generation. Management of Data(SIGMOD00), 2000:1-12.
9 张秋凤.加权关联规则挖掘算法研究及应用.天津:天津师范大学硕士研究生学位论文, 2011.
10 Cai C H, Ada W C F U, Cheng C H, et a1. Mining association rules with weighted items. In Proc of the International Database Engineering and Applications Symposium, Cardiff:Wales, 1998:68-77.
11 刘成海,危北海,姚树坤.肝硬化中西医结合诊疗共识.中国中西医结合消化杂志, 2011(4):277-279.
12 盛国光,黄育华,李晓东,等.慢性乙型肝炎中医辨证规律探讨.中西医结合肝病杂志, 2012, 12(6):382.
2 田瑾.基于复杂网络及关联规则的失眠用药中医临床数据挖掘研究.北京:北京中医药大学硕士研究生学位论文, 2015.
3 仝小林.论辨症、辨病、审因与辨证论治在临床中的应用.中医杂志,2013(2):93-95.
4 谢宁,郑洪新,赵文静.中医学基础.北京:中国中医药出版社, 2011(2):23-26.
5 汤尔群,黄玉燕,桑希生.《黄帝内经》组方理论探讨.安徽中医学院学报, 2012(1):1-3.
6 Brin S, Motwani R, Silverstein C. Beyond market basket:Generalizing association rules to correlations. Management of Data(SIGMOD97),1997:265-276.
7 韩家炜, Kamber M.数据挖掘概念与技术.北京:机械工业出版社,2012(7):55-61.
8 Han J, Pei J, Yin Y. Mining frequent patterns without candidate generation. Management of Data(SIGMOD00), 2000:1-12.
9 张秋凤.加权关联规则挖掘算法研究及应用.天津:天津师范大学硕士研究生学位论文, 2011.
10 Cai C H, Ada W C F U, Cheng C H, et a1. Mining association rules with weighted items. In Proc of the International Database Engineering and Applications Symposium, Cardiff:Wales, 1998:68-77.
11 刘成海,危北海,姚树坤.肝硬化中西医结合诊疗共识.中国中西医结合消化杂志, 2011(4):277-279.
12 盛国光,黄育华,李晓东,等.慢性乙型肝炎中医辨证规律探讨.中西医结合肝病杂志, 2012, 12(6):382.