cluster by market, using c4.5 and knn, not working...

expedia.cluster_by_market.r

length(unique(train$hotel_market))
hotel_markets<-unique(train$hotel_market)
market<-train[hotel_market==hotel_markets[2],]
summary(market)
unique(market$hotel_continent)

setkey(market)

market$date_time<-as.POSIXct(market$date_time)
market<-market[-which(market$srch_ci==""),]
market$srch_ci<-as.Date(market$srch_ci)
market$srch_co<-as.Date(market$srch_co)


market[['stay_time']]<-market$srch_co-market$srch_ci


market$stay_time<-as.numeric(market$stay_time)
market$hotel_cluster<-as.factor(market$hotel_cluster)
market$is_booking<-as.factor(market$is_booking)
market$site_name<-as.factor(market$site_name)
market$posa_continent<-as.factor(market$posa_continent)
market$user_location_city<-as.factor(market$user_location_city)
market$user_location_country<-as.factor(market$user_location_country)
market$user_location_region<-as.factor(market$user_location_region)
market$user_id<-as.factor(market$user_id)
market$is_mobile<-as.factor(market$is_mobile)
market$is_package<-as.factor(market$is_package)
market$channel<-as.factor(market$channel)
market$srch_destination_id<-as.factor(market$srch_destination_id)
market$srch_destination_type_id<-as.factor(market$srch_destination_type_id)
market$hotel_continent<-as.factor(market$hotel_continent)
market$hotel_country<-as.factor(market$hotel_country)
market$hotel_market<-as.factor(market$hotel_market)
market$hotel_country<-as.factor(market$hotel_country)
market$hotel_country<-as.factor(market$hotel_country)



summary(market)

#set training 
smp_size <- floor(0.70 * nrow(market))
## set the seed to make your partition reproductible
set.seed(123)
train_idx <- sample(seq_len(nrow(market)), size = smp_size)
train <- market[train_idx, ]
test <- market[-train_idx, ]

train$orig_destination_distance<-NULL
test$orig_destination_distance<-NULL

cls<-train$hotel_cluster
library(class)
knn(train,test,cls)



summary(time)
library(rpart)
fit <- rpart(hotel_cluster ~  date_time+stay_time+is_booking,
             method="class", data=samp1)

plot(fit)