##splits samples into Training set, Test set and Validation set in different ways.
#for c=1,2,3,4
## separates numbers in SequenceType strig variable
## Paste the folder with data <data\\c>  
DefectiveSamples<-read.csv("DescriptiveAnalysisDefectiveST.csv", sep=";")
ApplicationNames<-DefectiveSamples$Application
## convert values in faulty and non faulty, removeUsernumber, and move Failurenumber to the end
m<-length(ApplicationNames)

for(i in 1:4){ 
    directory<-paste("data\\c",i, sep="")
    D<- dir(directory, full.names=TRUE)
    H<- dir(directory, full.names=FALSE)
    for(j in 1:m){    
         data <- read.csv(D[j], header=TRUE, sep=",")
         n<-nrow(data)
         ##creates a vector with one "2" and all "1" for replication
         c<-c(1:n)
         d<-c-c+1
         d[1]<-2
         dataTemp<-rep(data$SequenceType,d)
         file<-paste("data\\c",i,"\\Temp_",H[j], sep="") 
         write.table(dataTemp, row.names=FALSE,col.names=FALSE,append=FALSE,file=file, sep=",",quote=FALSE)
         add<-read.csv(file,header=TRUE,sep=",")
         k<-ncol(add)
         colnames(add)<-paste("m",c(1:k), sep="")
         data1<-data[,-c(1)]
         merge<-cbind(add, data1)
         oldFile<-paste("C:\\Users\\brusso\\Desktop\\Pedrycz\\data\\c",i,"\\",H[j], sep="")
         finalFile<-paste("C:\\Users\\brusso\\Desktop\\Pedrycz\\data\\c",i,"\\New_",H[j], sep="")
         write.table(merge, row.names=FALSE,col.names=TRUE,append=FALSE,file=finalFile, sep=",",quote=FALSE)
         file.remove(file)
         file.remove(oldFile)
    }
}

for(i in 1:4){
    for(j in 1:m){
         directory<-paste("data\\c",i, sep="")
         D<- dir(directory, full.names=TRUE)
         H<- dir(directory, full.names=FALSE)
         file<-paste("data\\c",i,"\\Temp_",H[j], sep="") 
         oldFile<-paste("data\\c",i,"\\",H[j], sep="")
         file.remove(file)
         file.remove(oldFile)
    }
}