library("openxlsx")
library("C50")
library("reshape2")


dataCreditRating <- read.xlsx(xlsxFile="https://storage.googleapis.com/dqlab-dataset/credit_scoring_dqlab.xlsx")


dataCreditRating$risk_rating <- as.factor(dataCreditRating$risk_rating)
input_columns <-c("durasi_pinjaman_bulan", "jumlah_tanggungan")
datafeed <- dataCreditRating[,input_columns]


set.seed(100)
indeks_training_set <- sample(1:nrow(dataCreditRating),800)


input_training_set <- datafeed[indeks_training_set,]
class_training_set <- dataCreditRating[indeks_training_set,]$risk_rating
input_testing_set <- datafeed[-indeks_training_set,]


risk_rating_model <- C5.0(input_training_set, class_training_set,control=C5.0Control(label="Risk_Rating"))


input_testing_set$risk_rating <- dataCreditRating[-indeks_training_set,]$risk_rating
input_testing_set$hasil_prediksi <- predict(risk_rating_model,input_testing_set)
										   
result <- melt(input_testing_set[c("risk_rating","hasil_prediksi")],id.vars="risk_rating")	

head(result)
dcast(data=input_testing_set, hasil_prediksi ~risk_rating)