Full text link: tecdat.cn/?p=30605

Recently, we were asked by a client to write a research report on Chinese medicine patent compound treatment, including some graphics and statistical output.

Application of data mining techniques such as association rules and clustering methods to analyze the compatibility rules of traditional Chinese medicine patent compound prescriptions

Methods The patent compound prescriptions of traditional Chinese medicine were retrieved, and the compound prescriptions of external Chinese medicine and combination of Chinese and Western medicine were excluded. We were recently asked to write a research report on medication regularity, including some graphical and statistical output. The selected traditional Chinese medicine patent compounds were processed by terminology standardization, information was extracted, tables were established, data analysis software R was used to analyze data association rules, and network analysis software was used to perform cluster analysis.

view data

Convert to binary matrix data

colnames(data) <- paste0("X",1:ncol(data))

database <- NULL
for(i in 1:nrow(data)) {
  tmp <- integer(length(total_types))
复制代码

build apriori

plot(all_rules, method = "graph")
复制代码

Click on the title to view past content

R language uses association rules and clustering models to mine prescription data to explore the rules in drug compatibility

Swipe left and right to see more

Analysis of association rules of traditional Chinese medicine patent compound prescription pairs

The medicine pair is the basic form of prescription compatibility, which reflects the compatibility relationship among traditional Chinese medicines such as complementing each other, opposite complementing each other, and similar complementing each other. The Chinese medicines in the medicine pair have the characteristics of appearing in the prescription at the same time when the prescription is combined. Therefore, in the analysis of association rules, the medicine pair can be obtained by analyzing the rules with high confidence and two-way association.

Filter strong association rules based on confidence and support

K-means mean network cluster analysis

A complex network of compatibility relationships is formed among the drugs in the patent compound of traditional Chinese medicine for depression. Association rule analysis can be used to find drug pairs and strong association rules. Emergence makes it difficult to analyze the compatibility law, and the application of network clustering method can effectively find the compatibility law.

#聚类类别号
kmod$cluster
复制代码

View strong association rules in each category

Cluster 1

Cluster 2

The cluster analysis results of the compatibility relationship network showed the "communities" of traditional Chinese medicines commonly used in the treatment of depression, which reflected some combination of traditional Chinese medicines with relatively close and fixed compatibility in the compound prescription, and clinical application can improve the curative effect.

Click "Read the original text" at the end of the article

Get the full text and complete code data materials.

This article is selected from "R Language APRIORI Association Rules, K-MEANS Mean Value Clustering Data Mining Network Visualization of Medication Rules of Traditional Chinese Medicine Patent Compound Treatment".