China medicine has a long history, Chinese traditional medicine (TCM) is an important component of Chinese medicine, but also reflect the basic means of TCM curative effects. Compounds of TCM are mixture of drugs according to certain compatibility rules, complex composition. In clinical application certain herbs are selected following the compatibility law as a means, to achieve treatment for different diseases. Therefore, the effective components of Chinese medicine, the effective components of single drug or effective ingredients of the traditional Chinese medicine prescription are not compatibility, studying the superposition, inhibition the interaction between the components is essential. In recent years, the application of modern information technology has been developed, and applied in a lot of the research of TCM. Especially the technology of data mining, excavation of the Chinese native medicine compound prescription has become an important means of modernization of traditional Chinese medicine. Data mining technology is a set of scientific method to solve the problem of a large number of data from the application of various fields and put forward, mainly in order to discover the hidden knowledge and rules in the data, provide data and theoretical support for the decision scheme of human. Many TCM databases have been established. So we attempt to study the application of data mining technology in traditional Chinese medicine effective constituents in compound prescription and regulation, try to analyze and explore the effectiveness of compositions of TCM by computer. This paper uses the data mining technology as a means, analysis of composition rules and active components on the spleen and stomach and heat compound prescriptions, which mainly includes the following work: Data mining is the core algorithm. Some set of frequent items mining method based on APriori method, candidate set generation needs to constantly scanning the database, time consumption. FP-growth is a kind of mining method without generating candidate itemsets important frequent item sets.In this paper, based on the improved algorithm, the new algorithm uses the modified FP-tree and the head table structure, produce only FP-tree once, and produced only head table structure at each recursion. The new algorithm can obtain the same as the original algorithm of frequent item sets mining results, but the speed of at least twice faster than the FP-growth. Screening of the data, the other agent data preprocessing, create the database, use the improved algorithm, the initial establishment of effective component and computer composing principle of traditional Chinese medicine compound simulation system, the system through the data mining techniques: frequent item sets, association rules, clustering analysis algorithm, carries on the analysis to the frequency of drug ingredients, compound symptom frequency, medicine sickness connection, pharmaceutical composition clustering, find out the related effective ingredients and disorder of spleen and stomach and Qingre prescription among.
Author(s) Details:
Xiyuan Zhang,
Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, China.
Jianshe Yang,
Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, China, Tongji University School of Medicine, Shanghai 200072, China and Gansu Medical College, Pingliang 744000, China.
Please see the link here: https://stm.bookpi.org/DTACSC/article/view/9555
