Data description
Data comes from kaggle. It consists of information about reservations in certain hotel. Aim is to find which customers are likely to cancel reservations.
library(FactoMineR)
library(corrplot)
library(smacof)
library(clusterSim)
library(tidyverse)
library(ggplot2)
library(dplyr)
library(factoextra)
library(Rtsne)Used libraries:
FactoMineR: A library for multivariate exploratory
data analysis and factor analysis.
corrplot: A library for visualizing and analyzing
correlations between variables.
smacof: A library for multidimensional scaling.
clusterSim: A library for simulating datasets for
cluster analysis.
dplyr: A library for data manipulation and
analysis.
ggplot2: A library for creating static and interactive
visualizations.
tidyverse: A meta-library for data manipulation,
analysis, and visualization.
factoextra: A library for factorial analysis and
clustering with additional functionalities.
Rtsne: A library for fast and efficient implementation
of t-SNE in R
EDA
Read data:
data <- read.csv("reservations.csv", header = T)Names of the columns:
colnames(data)## [1] "Booking_ID"
## [2] "no_of_adults"
## [3] "no_of_children"
## [4] "no_of_weekend_nights"
## [5] "no_of_week_nights"
## [6] "type_of_meal_plan"
## [7] "required_car_parking_space"
## [8] "room_type_reserved"
## [9] "lead_time"
## [10] "arrival_year"
## [11] "arrival_month"
## [12] "arrival_date"
## [13] "market_segment_type"
## [14] "repeated_guest"
## [15] "no_of_previous_cancellations"
## [16] "no_of_previous_bookings_not_canceled"
## [17] "avg_price_per_room"
## [18] "no_of_special_requests"
## [19] "booking_status"Data types:
str(data)## 'data.frame': 36275 obs. of 19 variables:
## $ Booking_ID : chr "INN00001" "INN00002" "INN00003" "INN00004" ...
## $ no_of_adults : int 2 2 1 2 2 2 2 2 3 2 ...
## $ no_of_children : int 0 0 0 0 0 0 0 0 0 0 ...
## $ no_of_weekend_nights : int 1 2 2 0 1 0 1 1 0 0 ...
## $ no_of_week_nights : int 2 3 1 2 1 2 3 3 4 5 ...
## $ type_of_meal_plan : chr "Meal Plan 1" "Not Selected" "Meal Plan 1" "Meal Plan 1" ...
## $ required_car_parking_space : int 0 0 0 0 0 0 0 0 0 0 ...
## $ room_type_reserved : chr "Room_Type 1" "Room_Type 1" "Room_Type 1" "Room_Type 1" ...
## $ lead_time : int 224 5 1 211 48 346 34 83 121 44 ...
## $ arrival_year : int 2017 2018 2018 2018 2018 2018 2017 2018 2018 2018 ...
## $ arrival_month : int 10 11 2 5 4 9 10 12 7 10 ...
## $ arrival_date : int 2 6 28 20 11 13 15 26 6 18 ...
## $ market_segment_type : chr "Offline" "Online" "Online" "Online" ...
## $ repeated_guest : int 0 0 0 0 0 0 0 0 0 0 ...
## $ no_of_previous_cancellations : int 0 0 0 0 0 0 0 0 0 0 ...
## $ no_of_previous_bookings_not_canceled: int 0 0 0 0 0 0 0 0 0 0 ...
## $ avg_price_per_room : num 65 106.7 60 100 94.5 ...
## $ no_of_special_requests : int 0 1 0 0 0 1 1 1 1 3 ...
## $ booking_status : chr "Not_Canceled" "Not_Canceled" "Canceled" "Canceled" ...Data summary:
summary(data)## Booking_ID no_of_adults no_of_children no_of_weekend_nights
## Length:36275 Min. :0.000 Min. : 0.0000 Min. :0.0000
## Class :character 1st Qu.:2.000 1st Qu.: 0.0000 1st Qu.:0.0000
## Mode :character Median :2.000 Median : 0.0000 Median :1.0000
## Mean :1.845 Mean : 0.1053 Mean :0.8107
## 3rd Qu.:2.000 3rd Qu.: 0.0000 3rd Qu.:2.0000
## Max. :4.000 Max. :10.0000 Max. :7.0000
## no_of_week_nights type_of_meal_plan required_car_parking_space
## Min. : 0.000 Length:36275 Min. :0.00000
## 1st Qu.: 1.000 Class :character 1st Qu.:0.00000
## Median : 2.000 Mode :character Median :0.00000
## Mean : 2.204 Mean :0.03099
## 3rd Qu.: 3.000 3rd Qu.:0.00000
## Max. :17.000 Max. :1.00000
## room_type_reserved lead_time arrival_year arrival_month
## Length:36275 Min. : 0.00 Min. :2017 Min. : 1.000
## Class :character 1st Qu.: 17.00 1st Qu.:2018 1st Qu.: 5.000
## Mode :character Median : 57.00 Median :2018 Median : 8.000
## Mean : 85.23 Mean :2018 Mean : 7.424
## 3rd Qu.:126.00 3rd Qu.:2018 3rd Qu.:10.000
## Max. :443.00 Max. :2018 Max. :12.000
## arrival_date market_segment_type repeated_guest
## Min. : 1.0 Length:36275 Min. :0.00000
## 1st Qu.: 8.0 Class :character 1st Qu.:0.00000
## Median :16.0 Mode :character Median :0.00000
## Mean :15.6 Mean :0.02564
## 3rd Qu.:23.0 3rd Qu.:0.00000
## Max. :31.0 Max. :1.00000
## no_of_previous_cancellations no_of_previous_bookings_not_canceled
## Min. : 0.00000 Min. : 0.0000
## 1st Qu.: 0.00000 1st Qu.: 0.0000
## Median : 0.00000 Median : 0.0000
## Mean : 0.02335 Mean : 0.1534
## 3rd Qu.: 0.00000 3rd Qu.: 0.0000
## Max. :13.00000 Max. :58.0000
## avg_price_per_room no_of_special_requests booking_status
## Min. : 0.00 Min. :0.0000 Length:36275
## 1st Qu.: 80.30 1st Qu.:0.0000 Class :character
## Median : 99.45 Median :0.0000 Mode :character
## Mean :103.42 Mean :0.6197
## 3rd Qu.:120.00 3rd Qu.:1.0000
## Max. :540.00 Max. :5.0000Fix data types:
data$type_of_meal_plan <- as.factor(data$type_of_meal_plan)
data$room_type_reserved <- as.factor(data$room_type_reserved)
data$market_segment_type <- as.factor(data$market_segment_type)
data$booking_status <- as.factor(data$booking_status)Select numeric columns for PCA:
pca_cols <-
c(
"no_of_adults",
"no_of_children",
"no_of_weekend_nights",
"no_of_week_nights",
"required_car_parking_space",
"arrival_year",
"arrival_month",
"arrival_date",
"repeated_guest",
"no_of_previous_cancellations",
"no_of_previous_bookings_not_canceled",
"avg_price_per_room",
"no_of_special_requests"
)
data.pca <- data[, pca_cols]Show correlations:
corr.df<-cor(data.pca, method="pearson")
corrplot(corr.df, order ="alphabet", tl.cex=0.6)
Scale the data
data.pca <- data.Normalization(data.pca, type="n1", normalization="column")
data.pca <- data.pca[complete.cases(data.pca), ]PCA
Explained variances
sample.pca <- prcomp(data.pca)
fviz_screeplot(sample.pca)
Contributions of certain variables
fviz_contrib(sample.pca, choice = "var")
MDS
Calculate dissimilarities – distance matrix
dist.sub<-dist(t(data.pca))Run MDS
mds.object <-mds(dist.sub, ndim=2, type="ratio") Stress Decomposition Chart
plot(mds.object, plot.type = "stressplot") 
MDS for selected variables
plot(mds.object, pch=21, cex=as.numeric(mds.object$spp), bg="coral2", main="MDS for selected variables")
t-SNE
Perform t-SNE
tsne.obj <- Rtsne(data.pca, check_duplicates = FALSE)Prepare data for visualization
tnse.data <- data %>%
mutate(id = row_number())
tsne_results <- tsne.obj$Y %>%
as.data.frame() %>%
mutate(id=row_number())
tnse.data <- tnse.data %>%
inner_join(tsne_results, by="id")Vizualize results
tnse.data %>%
ggplot(aes(x=V1, y=V2, color=booking_status)) + geom_point() 
Conclusions
In conclusion, dimensionality reduction techniques were used to explore the Hotel Reservations Classification Dataset from Kaggle. The goal was to determine which customers are likely to cancel their reservations. To achieve this, several R libraries were utilized, including FactoMineR, corrplot, smacof, clusterSim, tidyverse, ggplot2, dplyr, factoextra, and Rtsne. Exploratory data analysis was performed by reading and summarizing the data, fixing the data types, and selecting numeric columns for principal component analysis (PCA). The PCA was then performed to determine the explained variances and contributions of certain variables, and multidimensional scaling (MDS) was performed to calculate dissimilarities and prepare visualizations. Finally, t-SNE was performed to further reduce the dimensions and prepare the data for visualization.