输入arulesViz的外部规则

时间:2015-03-08 10:55:20

标签: arules

我有一个excel文件规则:一个用于前因的列,另一个用于Consequents,另一个用于支持等等

我试图创建一个"规则"使用arulesViz绘制它们的对象。

http://www.inside-r.org/packages/cran/arules/docs/rhs

new("rules", ...)

我试图创建老虎机

lhs:Object of class itemMatrix; the left-hand-sides of the rules (antecedents)
rhs:Object of class itemMatrix; the right-hand-sides of the rules (consequents)
quality:a data.frame

要创建itemMatrix,我需要回到稀疏矩阵,但我认为这不会起作用,是否有某种方法可以导入"导入" arulesViz的规则?

Excel文件:

Antec   Conseq      Supp      Conf
MMMMAAA MMAAAA  0.061945    0.5
MMM,MA  MMAAAA  0.071944    0.6
MMMMAAA MMAAA   0.053948    0.5
MMM,MA  MMAAA   0.054948    0.7
AAAAAA  AAAA    0.090909    0.5

1 个答案:

答案 0 :(得分:1)

一种方法是为规则创建PMML文件并使用read.PMML。从头开始创建规则对象有点棘手。这是一个例子:

library("arules")
l <- list(c("MMMMAAA"), c("MMM", "MA"), "MMMMAAA", c("MMM","MA"), "AAAAA")
r <- list("MMAAAA", "MMAAAA", "MMAAA", "MMAAA", "AAAA")
q <- data.frame(
     support = c(0.061945, 0.071944, 0.053948, 0.054948,0.090909), 
     confidence = c(.5, .6, .5, .7, .5),
     lift = c(1,1,1,1,1)
)
### Note that I also added lift since arulesViz uses 
### lift in some visualizations.


### find unique item labels
items <- unique(c(unlist(l), unlist(r)))

### encode data as a rules object
r <- new("rules", lhs = encode(l, items), 
   rhs = encode(r, items), quality = q)

inspect(r)

### use a visualization as a scatter plot.
library("arulesViz")
plot(r)

希望这有帮助!