【微信小编】关联规则图

今天终于知道以前看到的关联规则图是怎么画出来的了。

效果图如下：

setwd("D://test//")
testdt<-read.csv("test.csv",header = TRUE,sep="\t")
head(testdt)
trans5<- as(split(testdt[,"disease_tp"], testdt[,"zjhm"]), "transactions")
inspect(trans5)
rules <- apriori(trans5, parameter = list(supp = 0.01, conf = 0.8, target = "rules",minlen=2))
summary(rules)

inspect(rules[1:20])

ordered_rules <- sort(rules, by="lift")
inspect(ordered_rules[1:20]) 

library(RColorBrewer)
library(arulesViz)
plot(rules, method="grouped",control=list(col = rev(brewer.pal(9, "Greens")[4:9])))  


plot(rules, measure="confidence", method="graph",control=list(type="items"),shading = "lift")

如果不要考虑前项后巷，只是单纯统计频繁的组合也可以用eclat

fsets <- eclat(trans5, parameter = list(support = 0.1,minlen=2)) 

inspect(fsets)

ordered_fsets <- sort(fsets, by="support")
inspect(ordered_fsets) 

出来的结果是这样的