R.如何确定规则

Question

Does anyone know how to identify rules that show which treatments are typically used for which diseases. I have this data. First column - patient, second - disease, third - medicine.

P1  D1  M1  
P1  D2  M1  
P2  D3  M2  M3  
P2  D4  M4  
P2  D1  M5  
P2  D2  M6  M7  M8  
P2  D1  M4  M9  
P2  D8  M10 
P3  D9  M11 

I read the following data with this code

t <- read.transactions("data.txt", format="basket", sep="\t", cols=1)
dt = apriori(t, parameter = list(support=0.002, confidence =0.5))
inspect(dt)

Answer 1

First, change your data from a wide to long form. You can use reshape() do this. Since you didn't provide variable names your code will look something like:

reshape(d, direction="long", varying=list(names(d)[2:7]), v.names="Treatment", idvar=c("PatientID"))

When you do that your data will look like:

P1  D1  M1
P1  D2  M1  
P2  D3  M2 
P2  D3  M3
P2  D4  M4  
P2  D1  M5  
P2  D2  M6
P2  D2  M7  
P2  D2  M8    
P2  D1  M4  
P2  D1  M9  
P2  D8  M10 
P3  D9  M11

After you do that you can easily create a 2x2 table to see frequency of treatments by diseases. For this your code will look like:

table(d$disease, d$M)

Producing (something like) this:

     M1 M2 M3
  D1  1  0  0
  D2  1  1  1