Bayes Theorem in R
I was thinking that we should do something useful right away. Bayes Theorem is a data algorithm for mining data. A good tutorial on the basic of this concept you can find here.
#reads in a table and creates a data frame that we call NewDisease NewDisease = read.table("c:/temp/NewDisease.txt", header = T, sep="\t") #creating the subsets Sick = subset(NewDisease, NewDisease=="YES") NotSick = subset(NewDisease, NewDisease=="NO") #sets the dimension of the dataframe dim(Sick) dim(NotSick) #a vector for obtaining the elements of the vector being sampled prob.Sick = colSums(Sick[,1:6]== "YES")/4 prob.NotSick = colSums(NotSick[,1:6]== "NO")/6 #if(!require("e1071")) install.packages("e1071") #Install the Naive Bayes module if needed uncomment the line below #install.packages("e1071", dependencies = TRUE) library(e1071) #the arguments on the left side takes values to be predicted with the predictor on the right side Classify = naiveBayes(NewDisease[1:10,1:6], NewDisease[1:10,7]) Classify #then we use the method predict predict(Classify, NewDisease[11,1:6])
`p(X,C) = p(C)prod((n),(k))p(X|C)`
# We can also set multiple variable values and then as long as we got corresponding values for the equation # what will do is divide the values in the print method v <- c( 2,5.5,6) t <- c(8, 3, 4) print(v/t)