The Study of the Compatibility Rules of Traditional Chinese Medicine Based on Apriori and HMETIS Hypergraph Partitioning Algorithm

Authors: Miao Wang, Jiayun Li, Li Chen, Yanjun Huang, Qiang Zhou, Lijuan Che and Huiliang Shang*

Abstract:

  Traditional Chinese Medicine (TCM) is the most important part of Traditional Chinese Culture. One of the concentration of studies carried by scholars of TCM is how to uncover core herbs for certain syndrome and manipulate the combination rules of herbs to increase the efficacy of treatments. Although TCM has established some combination rules of herbs and core herbs for certain syndrome, most of those rules are based on empirical analysis, which makes them hard to study. Ever since concepts of Big Data and machine learning have been brought forward, how to use the data mining algorithm to effectively discovery the core herbs and combination rules gradually becomes the main aspect of TCM informatics.

  In this paper, a hypergraph cluster detection algorithm based on Apriori has been applied in the clinical data about lung cancer. The result shows that all 15 traditional herbs found by the algorithm are close to the core concepts of famous doctors to strengthen the body resistance and to eliminate pathogenic factors. 

Keywords:

Data Mining, Combination Rules of Herbs, Apriori Algorithm, Hypergraph, Community Detection

Published. VLDB DMAH 2015, The First International Workshop on Data Management and Analytics for Medicine and Healthcare

Full text:

http://link.springer.com/chapter/10.1007/978-3-319-41576-5_2

Pages from (Lecture Notes in Computer Science 9579) Fusheng Wang, Gang Luo, Chunhua Weng, Arijit Khan, Prasenjit Mitra, Cong Yu (eds.) - Biomedical Data Management and Graph Online Querying_ VLDB 2015 Workshops,-2.pdf