#Clients matrix with data from clientes.txt file
PATH= "Desktop/Business Intelligence/Tareas/Tarea 2/clientes.txt"
clients=read.table(PATH, sep = " ")
#Matrix with only 15 first columns of the original matrix
names <- c("ClientSubtype", "HouseNum", "HouseInhabitantsNum", "AvgAge", "ClientType", "Religion", "CivilState", "FamilyType", "ScholarEducation", "JobType", "SocialClass", "HouseType", "CarsNum", "HealthType", "AvgIncome")
M<-clients[,1:15]
colnames(M) <- names

#Exercise 1 - Find relevant variables
stdM <- scale(M)
corM <- cor(stdM)
varM <- apply(M, 2, var)

#Remove variables and transform data into categorical values
M$ClientSubtype<-NULL #"ClientSubType" was removed due to its high correlation with "ClientType"
M$HouseNum<-factor(M$HouseNum, labels= c("1","2","3","4","5","6","7","8","10"))
M$HouseInhabitantsNum<-NULL #factor(M$HouseInhabitantsNum, labels= c("1","2","3","4","5"))
M$AvgAge<-factor(M$AvgAge, labels = c("20-30","30-40","40-50","50-60","60-70","70-80"))
M$ClientType<-factor(M$ClientType, labels = c("Successful hedonists", "Driven Growers","Average Family","Career Loners","Living well","Cruising Seniors","Retired and Religeous","Family with grown ups","Conservative families","Farmers"))
M$Religion<-factor(M$Religion, labels = c("Catholic","Protestant","Other","No religion"))
M$CivilState<-factor(M$CivilState, labels = c("Married", "cohabitation", "Others"))
M$FamilyType<-factor(M$FamilyType, labels = c("Single", "Family without children", "Family with children"))
M$ScholarEducation<-NULL #factor(M$ScholarEducation, labels = c("High", "Mid", "Low"))
M$JobType<-factor(M$JobType, labels = c("High position", "Enterprising", "Farmer","Middle management", "Qualified worker", "Unqualified worker"))
M$SocialClass<-factor(M$SocialClass, labels = c("A", "B1", "B2", "C", "D"))
M$HouseType<-factor(M$HouseType, labels = c("Tenant", "Owner"))
M$CarsNum<-factor(M$CarsNum, labels = c("One", "Two", "No car"))
M$HealthType<-factor(M$HealthType, labels = c("Fonasa", "Isapre"))
M$AvgIncome<-factor(M$AvgIncome, labels = c("Low", "Mid Low", "Medium", "Mid High", "High"))