EDA of Hotel Booking Dataset - Kaggle

Mehedi Azad
Data Analyst | Data Science | Machine Learning |

AI | Power BI | Python | Data nerd
https://www.linkedin.com/in/mehediazad/
In [ ]: #linkedin link: https://www.linkedin.com/in/mehediazad/

# dataset link: https://drive.google.com/drive/folders/1NPxYfyBHfHOoPpHLd9qbkfFWA
In [1]: import numpy as np

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
# import plotly.express as px

In [2]: df = pd.read_csv(r'G:\dataset\EDA ofhotel booking -kaggle\hotel_bookings.csv\hote

df
Out[2]:
hotel is_canceled lead_time arrival_date_year arrival_date_month arrival_date_week_num
Resort
0 0 342 2015 July
Hotel
Resort
1 0 737 2015 July
Hotel
Resort
2 0 7 2015 July
Hotel
Resort
3 0 13 2015 July
Hotel
Resort
4 0 14 2015 July
Hotel
... ... ... ... ... ...
City
119385 0 23 2017 August
Hotel
City
119386 0 102 2017 August
Hotel
City
119387 0 34 2017 August
Hotel
City
119388 0 109 2017 August
Hotel
City
119389 0 205 2017 August
Hotel
119390 rows × 32 columns
Checking First 5 records from this dataset using head()

function
In [3]: df.head()
Out[3]:
hotel is_canceled lead_time arrival_date_year arrival_date_month arrival_date_week_number
Resort
0 0 342 2015 July 27
Hotel
Resort
1 0 737 2015 July 27
Hotel
Resort
2 0 7 2015 July 27
Hotel
Resort
3 0 13 2015 July 27
Hotel
Resort
4 0 14 2015 July 27
Hotel
Checking Last 5 records from this dataset using tail()

function
In [4]: df.tail()
Out[4]:
hotel is_canceled lead_time arrival_date_year arrival_date_month arrival_date_week_numb
City
119385 0 23 2017 August
Hotel
City
119386 0 102 2017 August
Hotel
City
119387 0 34 2017 August
Hotel
City
119388 0 109 2017 August
Hotel
City
119389 0 205 2017 August
Hotel
Checking columns of this dataset

In [5]: df.columns
Out[5]: Index(['hotel', 'is_canceled', 'lead_time', 'arrival_date_year',
'arrival_date_month', 'arrival_date_week_number',
'arrival_date_day_of_month', 'stays_in_weekend_nights',
'stays_in_week_nights', 'adults', 'children', 'babies', 'meal',
'country', 'market_segment', 'distribution_channel',
'is_repeated_guest', 'previous_cancellations',
'previous_bookings_not_canceled', 'reserved_room_type',
'assigned_room_type', 'booking_changes', 'deposit_type', 'agent',
'company', 'days_in_waiting_list', 'customer_type', 'adr',
'required_car_parking_spaces', 'total_of_special_requests',
'reservation_status', 'reservation_status_date'],
dtype='object')
Showing Statistical information regarding the data using

describe() function
In [6]: df.describe()
Out[6]:
is_canceled lead_time arrival_date_year arrival_date_week_number arrival_date_day_
count 119390.000000 119390.000000 119390.000000 119390.000000 1193
mean 0.370416 104.011416 2016.156554 27.165173
std 0.482918 106.863097 0.707476 13.605138
min 0.000000 0.000000 2015.000000 1.000000
25% 0.000000 18.000000 2016.000000 16.000000
50% 0.000000 69.000000 2016.000000 28.000000
75% 1.000000 160.000000 2017.000000 38.000000
max 1.000000 737.000000 2017.000000 53.000000
Check the informations about the dataset using info()

function
In [7]: df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 119390 entries, 0 to 119389
Data columns (total 32 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 hotel 119390 non-null object
1 is_canceled 119390 non-null int64
2 lead_time 119390 non-null int64
3 arrival_date_year 119390 non-null int64
4 arrival_date_month 119390 non-null object
5 arrival_date_week_number 119390 non-null int64
6 arrival_date_day_of_month 119390 non-null int64
7 stays_in_weekend_nights 119390 non-null int64
8 stays_in_week_nights 119390 non-null int64
9 adults 119390 non-null int64
10 children 119386 non-null float64
11 babies 119390 non-null int64
12 meal 119390 non-null object
13 country 118902 non-null object
14 market_segment 119390 non-null object
15 distribution_channel 119390 non-null object
16 is_repeated_guest 119390 non-null int64
17 previous_cancellations 119390 non-null int64
18 previous_bookings_not_canceled 119390 non-null int64
19 reserved_room_type 119390 non-null object
20 assigned_room_type 119390 non-null object
21 booking_changes 119390 non-null int64
22 deposit_type 119390 non-null object
23 agent 103050 non-null float64
24 company 6797 non-null float64
25 days_in_waiting_list 119390 non-null int64
26 customer_type 119390 non-null object
27 adr 119390 non-null float64
28 required_car_parking_spaces 119390 non-null int64
29 total_of_special_requests 119390 non-null int64
30 reservation_status 119390 non-null object
31 reservation_status_date 119390 non-null object
dtypes: float64(4), int64(16), object(12)
memory usage: 29.1+ MB
Checking number of rows & columns in the dataset using

shape
In [8]: df.shape
# There is 119390 rows & 32 columns
Out[8]: (119390, 32)
Checking the data types of the dataset using dtypes

In [9]: df.dtypes
Out[9]: hotel object
is_canceled int64
lead_time int64
arrival_date_year int64
arrival_date_month object
arrival_date_week_number int64
arrival_date_day_of_month int64
stays_in_weekend_nights int64
stays_in_week_nights int64
adults int64
children float64
babies int64
meal object
country object
market_segment object
distribution_channel object
is_repeated_guest int64
previous_cancellations int64
previous_bookings_not_canceled int64
reserved_room_type object
assigned_room_type object
booking_changes int64
deposit_type object
agent float64
company float64
days_in_waiting_list int64
customer_type object
adr float64
required_car_parking_spaces int64
total_of_special_requests int64
reservation_status object
reservation_status_date object
dtype: object
Checking null values present in each column using

isnull().sum()
In [10]: df.isnull().sum()
Out[10]: hotel 0
is_canceled 0
lead_time 0
arrival_date_year 0
arrival_date_month 0
arrival_date_week_number 0
arrival_date_day_of_month 0
stays_in_weekend_nights 0
stays_in_week_nights 0
adults 0
children 4
babies 0
meal 0
country 488
market_segment 0
distribution_channel 0
is_repeated_guest 0
previous_cancellations 0
previous_bookings_not_canceled 0
reserved_room_type 0
assigned_room_type 0
booking_changes 0
deposit_type 0
agent 16340
company 112593
days_in_waiting_list 0
customer_type 0
adr 0
required_car_parking_spaces 0
total_of_special_requests 0
reservation_status 0
reservation_status_date 0
dtype: int64
In [11]: df.head()
Out[11]:
Resort
0 0 342 2015 July 27
Hotel
Resort
1 0 737 2015 July 27
Hotel
Resort
2 0 7 2015 July 27
Hotel
Resort
3 0 13 2015 July 27
Hotel
Resort
4 0 14 2015 July 27
Hotel
Droping 'company' column because it contains a large
amount of null values
In [12]: df = df.drop('company', axis=1)

In [13]: df.shape
Out[13]: (119390, 31)
In [14]: df.columns
'days_in_waiting_list', 'customer_type', 'adr',
dtype='object')
there is a large amount of null values in 'agent' column.

replacing null values with zero using fillna() function
In [15]: df['agent'] = df['agent'].fillna(0)
In [16]: df['agent']
Out[16]: 0 0.0
1 0.0
2 0.0
3 304.0
4 240.0
...
119385 394.0
119386 9.0
119387 9.0
119388 89.0
119389 9.0
Name: agent, Length: 119390, dtype: float64
To check count number in each country using

'value_counts()' function
In [17]: df['country'].value_counts()
Out[17]: PRT 48590
GBR 12129
FRA 10415
ESP 8568
DEU 7287
...
ASM 1
BWA 1
HND 1
MLI 1
BHS 1
Name: country, Length: 177, dtype: int64
There is also some null values in 'country' column.

Replaing it with the most value count country 'PRT'
In [18]: # To replace the null values of country with prt
df['country'] = df['country'].fillna('PRT')
In [19]: df['country'].value_counts()
Out[19]: PRT 49078
GBR 12129
FRA 10415
ESP 8568
DEU 7287
...
ASM 1
BWA 1
HND 1
MLI 1
BHS 1
Name: country, Length: 177, dtype: int64
Now again Checking null values after clearing

them in each column using isnull().sum()
In [20]: df.isnull().sum()
# now there is no null values
Out[20]: hotel 0
is_canceled 0
lead_time 0
arrival_date_year 0
arrival_date_month 0
arrival_date_week_number 0
arrival_date_day_of_month 0
stays_in_weekend_nights 0
stays_in_week_nights 0
adults 0
children 4
babies 0
meal 0
country 0
market_segment 0
distribution_channel 0
is_repeated_guest 0
previous_cancellations 0
previous_bookings_not_canceled 0
reserved_room_type 0
assigned_room_type 0
booking_changes 0
deposit_type 0
agent 0
days_in_waiting_list 0
customer_type 0
adr 0
required_car_parking_spaces 0
total_of_special_requests 0
reservation_status 0
reservation_status_date 0
dtype: int64
To find correlation between each data using 'corr()'

function
In [21]: df.corr()
Out[21]:
is_canceled lead_time arrival_date_year arrival_date_week_numbe
is_canceled 1.000000 0.293123 0.016660 0.00814
lead_time 0.293123 1.000000 0.040142 0.12687
arrival_date_year 0.016660 0.040142 1.000000 -0.54056
arrival_date_week_number 0.008148 0.126871 -0.540561 1.00000
arrival_date_day_of_month -0.006130 0.002268 -0.000221 0.06680
stays_in_weekend_nights -0.001791 0.085671 0.021497 0.01820
stays_in_week_nights 0.024765 0.165799 0.030883 0.01555
adults 0.060017 0.119519 0.029635 0.02590
children 0.005048 -0.037622 0.054624 0.00551
babies -0.032491 -0.020915 -0.013192 0.01039
is_repeated_guest -0.084793 -0.124410 0.010341 -0.03013
previous_cancellations 0.110133 0.086042 -0.119822 0.03550
previous_bookings_not_canceled -0.057358 -0.073548 0.029218 -0.02090
booking_changes -0.144381 0.000149 0.030872 0.00550
agent -0.046529 -0.012640 0.056463 -0.01824
days_in_waiting_list 0.054186 0.170084 -0.056497 0.02293
adr 0.047557 -0.063077 0.197580 0.07579
required_car_parking_spaces -0.195498 -0.116451 -0.013684 0.00192
total_of_special_requests -0.234658 -0.095712 0.108531 0.02614
Type Markdown and LaTeX: 𝛼2

To check the number of duplicated values using
'duplicated().sum()'
In [22]: df.duplicated().sum()
Out[22]: 32020
To drop the duplicates permanently using

'drop_duplicates(inplace=True)' fnction
In [23]: df.drop_duplicates(inplace=True)
In [24]: df.shape
Out[24]: (87370, 31)
Cheking unique 'Hotel' column using 'unique()'

function
In [25]: df['hotel'].unique()
Out[25]: array(['Resort Hotel', 'City Hotel'], dtype=object)

Checking Total number of 'hotel'column using
'value_counts()' function
In [26]: df['hotel'].value_counts()
Out[26]: City Hotel 53426
Resort Hotel 33944
Name: hotel, dtype: int64
Count plot on single categorical variable 'hotel'

In [27]: sns.countplot(x='hotel', data=df)
#show the plot using 'plt.show()'
plt.show()
In [28]: df.columns
dtype='object')
To count the number of visitors canceled in Total

In [29]: df.groupby(['is_canceled'])['hotel'].value_counts().sum()
Out[29]: 87370
To anlyse which hotel gets cancelled the most by the

customers
In [30]: sns.barplot(x=df['hotel'], y=df['is_canceled'])
plt.show()

In [31]: df.groupby(['arrival_date_month'])['hotel'].count().sort_values(ascending=False)
Out[31]: arrival_date_month
August 11257
July 10055
May 8354
April 7905
June 7765
March 7510
October 6932
September 6689
February 6091
December 5128
November 4993
January 4691
To know the count the visitors in each month

In [32]: sns.countplot(x='arrival_date_month', data=df)
plt.xticks(rotation=90)
plt.rcParams['figure.figsize'] = (13, 6)
plt.show()
In [33]: df.head()
Out[33]:
Resort
0 0 342 2015 July 27
Hotel
Resort
1 0 737 2015 July 27
Hotel
Resort
2 0 7 2015 July 27
Hotel
Resort
3 0 13 2015 July 27
Hotel
Resort
4 0 14 2015 July 27
Hotel
To know the count the visitors in each week

In [34]: df.groupby(['arrival_date_week_number'])['hotel'].count()
Out[34]: arrival_date_week_number
1 862
2 945
3 1049
4 1124
5 1101
6 1295
7 1629
8 1523
9 1579
10 1630
11 1657
12 1572
13 1817
14 1693
15 1989
16 1736
17 1878
18 2087
19 1813
20 1843
21 2043
22 1753
23 1872
24 1746
25 1786
26 1739
27 2166
28 2343
29 2197
30 2335
31 2286
32 2449
33 2793
34 2491
35 2105
36 1626
37 1474
38 1634
39 1590
40 1427
41 1663
42 1445
43 1605
44 1548
45 1314
46 1141
47 1288
48 1199
49 1169
50 1053
51 785
52 1061
53 1422
In [35]: df.groupby(['hotel'])['lead_time'].count().sort_values(ascending=False)
Out[35]: hotel
City Hotel 53426
Resort Hotel 33944
Name: lead_time, dtype: int64
In [36]: df.groupby(['arrival_date_day_of_month'])['arrival_date_month'].count().sort_valu
Out[36]: arrival_date_day_of_month
17 3018
2 3015
26 3000
5 2980
16 2957
19 2949
12 2928
28 2924
18 2923
11 2915
20 2915
27 2902
29 2880
9 2878
15 2868
25 2837
3 2833
21 2822
13 2812
8 2808
6 2804
4 2798
10 2785
23 2776
24 2774
30 2770
1 2769
7 2704
14 2692
22 2601
31 1733
Name: arrival_date_month, dtype: int64

In [37]: sns.barplot(x=df['hotel'], y=df['lead_time'])
plt.show()
In [38]: df.groupby(['arrival_date_day_of_month'])['hotel'].count().sort_values(ascending=
Out[38]: arrival_date_day_of_month
17 3018
2 3015
26 3000
5 2980
16 2957
19 2949
12 2928
28 2924
18 2923
11 2915
20 2915
27 2902
29 2880
9 2878
15 2868
25 2837
3 2833
21 2822
13 2812
8 2808
6 2804
4 2798
10 2785
23 2776
24 2774
30 2770
1 2769
7 2704
14 2692
22 2601
31 1733

In [39]: sns.countplot(x='arrival_date_day_of_month', data=df)
plt.rcParams['figure.figsize'] = (13, 6)
plt.show()
In [40]: df.columns
dtype='object')
To count the number of meal

In [41]: df['meal'].count()
Out[41]: 87370
In [42]: data = df['meal'].value_counts()

print(data)
labels = df['meal'].unique()
BB 67955
SC 9481
HB 9084
Undefined 490
FB 360
Name: meal, dtype: int64
In [43]: explode = (0.1, 0.0, 0.2, 0.3, 0.7)

plt.figure(figsize =(10, 8))
plt.pie(data,labels=labels, explode=explode, autopct='%.2f')
plt.show()
In [44]: fig, (ax1, ax2) = plt.subplots(1,2, figsize=(12,6))

ax1.scatter(x=df['hotel'], y=df['meal'], marker='+', color='red')
ax2.scatter(x=df['hotel'], y=df['stays_in_weekend_nights'], marker='^', color='bl
Out[44]: <matplotlib.collections.PathCollection at 0x1a443345610>

In [45]: df.columns
dtype='object')
To count the number of adults in each hotel

In [46]: df.groupby(['hotel'])['adults'].sum()

Out[46]: hotel
City Hotel 100246
Resort Hotel 63656
Name: adults, dtype: int64
In [47]:
sns.scatterplot( data=df, x="hotel", y="adults", hue='meal')
plt.grid()
plt.show()

In [48]:
sns.scatterplot(data=df, x='hotel', y='children', hue='meal')
plt.grid()
plt.show()
In [49]: sns.scatterplot(data=df, x='hotel', y='babies', hue='meal')

plt.grid()
plt.show()
Insights:
1. Adult people prefer to come to Resort hotel and mostly order bb meal.
2. Most of the Childrens also prefer Resort hotel and mostly order bb meal.
3. The family with the babies prefer city hotel and the meal that is bb.
In [50]: df.columns
dtype='object')
In [51]: df['market_segment'].value_counts()
Out[51]: Online TA 51613
Offline TA/TO 13886
Direct 11798
Groups 4940
Corporate 4202
Complementary 702
Aviation 227
Undefined 2
Name: market_segment, dtype: int64
Market segment of both hotels

In [52]: sns.countplot(x='market_segment', data=df)
plt.rcParams['figure.figsize'] = (12,6)
plt.show()
In [53]: sns.countplot(x='market_segment', data=df, hue='hotel', palette='magma')
plt.rcParams['figure.figsize'] = (12,6)
plt.title('Market segment from both hotel')
plt.show()
checking customer ratio staying weekend night

In [54]: df.groupby(['hotel'])['stays_in_weekend_nights'].count().sort_values(ascending=Fa
Out[54]: hotel
City Hotel 53426
Resort Hotel 33944
Name: stays_in_weekend_nights, dtype: int64

In [55]: sns.countplot(x='stays_in_weekend_nights', data=df, hue='hotel')
# plt.rcParams['figure.figsize']=(12,6)
plt.figure(figsize=(12,6))
plt.show()
<Figure size 864x432 with 0 Axes>
Insight
Most of the people prefer to stay weekend night in City Hotel.
checking customer ratio staying week in night

In [56]: df['stays_in_week_nights'].count()
Out[56]: 87370
In [57]: df.groupby(['hotel'])['stays_in_week_nights'].count().sort_values(ascending=False
Out[57]: hotel
City Hotel 53426
Resort Hotel 33944
Name: stays_in_week_nights, dtype: int64
In [58]: sns.countplot(x='stays_in_week_nights',data=df, hue='hotel')

plt.show()
<Figure size 864x432 with 0 Axes>
Insight
Most of the people prefer to stay week in night in City Hotel.
Checking Reservation of hotel canceled in each hotel

In [59]: sns.countplot(x="market_segment", data=df, hue="is_canceled", palette="magma")
plt.title("Reservation cancelled from both hotels")
plt.show()

In [60]: list(df.columns)
Out[60]: ['hotel',
'is_canceled',
'lead_time',
'arrival_date_year',
'arrival_date_month',
'arrival_date_week_number',
'arrival_date_day_of_month',
'stays_in_weekend_nights',
'stays_in_week_nights',
'adults',
'children',
'babies',
'meal',
'country',
'market_segment',
'distribution_channel',
'is_repeated_guest',
'previous_cancellations',
'previous_bookings_not_canceled',
'reserved_room_type',
'assigned_room_type',
'booking_changes',
'deposit_type',
'agent',
'days_in_waiting_list',
'customer_type',
'adr',
'required_car_parking_spaces',
'total_of_special_requests',
'reservation_status',
'reservation_status_date']
In [61]: df.groupby(['market_segment','hotel']).count()
Out[61]:
is_canceled lead_time arrival_date_year arrival_date_month arrival_date_
market_segment hotel
Aviation City
227 227 227 227
Hotel
Complementary City
513 513 513 513
Hotel
Resort
189 189 189 189
Hotel
Corporate City
2227 2227 2227 2227
Hotel
Resort
1975 1975 1975 1975
Hotel
Direct City
5559 5559 5559 5559
Hotel
Resort
6239 6239 6239 6239
Hotel
Groups City
2635 2635 2635 2635
Hotel
Resort
2305 2305 2305 2305
Hotel
Offline TA/TO City

7271 7271 7271 7271
Hotel
Resort
6615 6615 6615 6615
Hotel
Online TA City
34992 34992 34992 34992
Hotel
Resort
16621 16621 16621 16621
Hotel
Undefined City
2 2 2 2
Hotel
In [62]: df.groupby(['market_segment'])['hotel'].count()
Out[62]: market_segment
Aviation 227
Complementary 702
Corporate 4202
Direct 11798
Groups 4940
Offline TA/TO 13886
Online TA 51613
Undefined 2
In [63]: sns.countplot(x='market_segment', data=df, hue='hotel', palette='magma')

plt.title('Market segment in each hotel')
plt.show()
Checking Repeated guest

In [64]: df['is_repeated_guest'].value_counts()
# 0 means there are no repeated guests
# 1 means there are repeated guests
Out[64]: 0 83955
1 3415
Name: is_repeated_guest, dtype: int64

In [65]: df.groupby(['is_repeated_guest'])['hotel'].value_counts()
# df.groupby(['hotel'])['is_repeated_guest'].value_counts()

# 0 means there are no repeated guests
# 1 means there are repeated guests
Out[65]: is_repeated_guest hotel
0 City Hotel 51718
Resort Hotel 32237
1 City Hotel 1708
Resort Hotel 1707
In [66]: df.groupby(['hotel','is_repeated_guest']).count()
Out[66]:
is_canceled lead_time arrival_date_year arrival_date_month arrival_dat
hotel is_repeated_guest
City 0 51718 51718 51718 51718

Hotel
1 1708 1708 1708 1708
Resort 0 32237 32237 32237 32237

Hotel
1 1707 1707 1707 1707
In [67]: df.groupby(['is_repeated_guest','hotel']).count()
Out[67]:
is_repeated_guest hotel
0 City
51718 51718 51718 51718
Hotel
Resort
32237 32237 32237 32237
Hotel
1 City
1708 1708 1708 1708
Hotel
Resort
1707 1707 1707 1707
Hotel
Out[68]: ['hotel',
'is_canceled',
'lead_time',
'adults',
'children',
'babies',
'meal',
'country',
'market_segment',
'booking_changes',
'deposit_type',
'agent',
'customer_type',
'adr',
In [69]: df['distribution_channel'].value_counts()
Out[69]: TA/TO 69133
Direct 12980
Corporate 5071
GDS 181
Undefined 5
Name: distribution_channel, dtype: int64

In [70]: sns.countplot(x='distribution_channel',data=df)
plt.show()
In [71]: sns.countplot(x='distribution_channel',data=df, hue='hotel')
plt.title('Distribution channel from both hotels')
plt.show()
In [72]: df['reservation_status'].value_counts()
Out[72]: Check-Out 63346
Canceled 23010
No-Show 1014
Name: reservation_status, dtype: int64
In [73]: df.groupby(['hotel', 'reservation_status']).count()
Out[73]:
hotel reservation_status
City Canceled 15301 15301 15301 15301

Hotel
Check-Out 37377 37377 37377 37377
No-Show 748 748 748 748
Resort Canceled 7709 7709 7709 7709

Hotel
Check-Out 25969 25969 25969 25969
No-Show 266 266 266 266
In [74]: sns.countplot(x='reservation_status', data=df, hue='hotel')
plt.show()
In [75]: sns.countplot(x='hotel', data=df, hue='reservation_status')

plt.show()
To count adr of each hotel

In [76]: df.groupby(['arrival_date_month'])['adr'].mean()
Out[76]: arrival_date_month
April 103.631776
August 150.876120
December 81.425918
February 74.731739
January 70.061422
July 135.521525
June 119.750120
March 81.624414
May 111.191058
November 72.768983
October 90.167276
September 112.081873
Name: adr, dtype: float64
Out[77]: ['hotel',
'is_canceled',
'lead_time',
'adults',
'children',
'babies',
'meal',
'country',
'market_segment',
'booking_changes',
'deposit_type',
'agent',
'customer_type',
'adr',
In [78]: #To show the adr in each month

sns.scatterplot(x=df['arrival_date_month'], y=df['adr'])
plt.show()
In [79]: df['reserved_room_type'].value_counts()
Out[79]: A 56530
D 17397
E 6047
F 2822
G 2052
B 999
C 915
H 596
P 6
L 6
Name: reserved_room_type, dtype: int64

In [80]: sns.countplot(x=df['reserved_room_type'],data=df)
plt.show()
In [81]: df['assigned_room_type'].value_counts()
Out[81]: A 46301
D 22422
E 7193
F 3625
G 2498
C 2165
B 1820
H 706
I 357
K 276
P 6
L 1
Name: assigned_room_type, dtype: int64

In [82]: sns.countplot(x=df['assigned_room_type'],data=df)
plt.show()
In [83]: df.groupby(['hotel'])['assigned_room_type'].value_counts()
Out[83]: hotel assigned_room_type
City Hotel A 33403
D 13209
E 2051
F 1970
B 1661
G 691
K 276
C 161
P 4
Resort Hotel A 12898
D 9213
E 5142
C 2004
G 1807
F 1655
H 706
I 357
B 159
P 2
L 1
Name: assigned_room_type, dtype: int64

Out[84]: ['hotel',
'is_canceled',
'lead_time',
'adults',
'children',
'babies',
'meal',
'country',
'market_segment',
'booking_changes',
'deposit_type',
'agent',
'customer_type',
'adr',
In [85]: df.groupby(['previous_cancellations'])['hotel'].value_counts()
Out[85]: previous_cancellations hotel
0 City Hotel 52248
Resort Hotel 33437
1 City Hotel 966
Resort Hotel 441
2 City Hotel 72
Resort Hotel 40
3 City Hotel 51
Resort Hotel 10
4 City Hotel 24
Resort Hotel 6
5 City Hotel 16
Resort Hotel 3
6 City Hotel 17
11 City Hotel 27
13 City Hotel 4
14 Resort Hotel 1
19 Resort Hotel 1
21 City Hotel 1
24 Resort Hotel 2
25 Resort Hotel 2
26 Resort Hotel 1
In [86]: df.groupby(['hotel'])['previous_cancellations'].value_counts()
Out[86]: hotel previous_cancellations
City Hotel 0 52248
1 966
2 72
3 51
11 27
4 24
6 17
5 16
13 4
21 1
Resort Hotel 0 33437
1 441
2 40
3 10
4 6
5 3
24 2
25 2
14 1
19 1
26 1
Name: previous_cancellations, dtype: int64

In [87]: sns.scatterplot(data=df, x=df['previous_cancellations'], y=df['customer_type'], h
plt.grid()
plt.show()
In [88]: df['customer_type'].value_counts()
Out[88]: Transient 71968
Transient-Party 11719
Contract 3139
Group 544
Name: customer_type, dtype: int64
Pie chart of customer_type

In [89]: data = df['customer_type'].value_counts()
print(data)
labels = df['customer_type'].unique()

explode = (0.1, 0.1, 0.2, 0.4)
plt.show()
Transient 71968
Transient-Party 11719
Contract 3139
Group 544
Name: customer_type, dtype: int64
In [90]: df['deposit_type'].value_counts()
Out[90]: No Deposit 86225
Non Refund 1038
Refundable 107
Name: deposit_type, dtype: int64
Pie chart of deposit_type

In [91]: data = df['deposit_type'].value_counts()
# print(data)
labels = df['deposit_type'].unique()

explode = (0.2, 0.3, 0.6)
plt.show()
countplot chart of deposit_type

In [92]: sns.countplot(x=df['deposit_type'], data=df, hue='hotel')
plt.show()

Out[93]: ['hotel',
'is_canceled',
'lead_time',
'adults',
'children',
'babies',
'meal',
'country',
'market_segment',
'booking_changes',
'deposit_type',
'agent',
'customer_type',
'adr',
To know the number of guest who have not cancelled the

booking
In [94]: # df[df['is_canceled']==0]['arrival_date_month'].value_counts()
month_guest = df[df['is_canceled']==0]['arrival_date_month'].value_counts().reset
month_guest.columns = ['month','Number of guest']
month_guest
Out[94]:
month Number of guest
0 August 7634
1 July 6857
2 May 5912
3 March 5680
4 April 5497
5 June 5411
6 October 5290
7 September 5047
8 February 4676
9 November 3939
10 December 3750
11 January 3653
To know the number of guests who have not cancelled the

booking each month.
In [95]:
sns.lineplot(data=month_guest, x ="month", y ="Number of guest")
plt.grid()
plt.show()
# To KNOW which visitors havent cancelled the booking

In [96]: data = df[df['is_canceled']==0]['arrival_date_month'].value_counts()
print(data)
labels = df['arrival_date_month'].unique()

# explode = (0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1, 1.2)
explode = (0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1)
plt.show()
August 7634
July 6857
May 5912
March 5680
April 5497
June 5411
October 5290
September 5047
February 4676
November 3939
December 3750
January 3653
Name: arrival_date_month, dtype: int64
To count the number of guest each year

In [97]: df['arrival_date_year'].value_counts()
Out[97]: 2016 42377
2017 31683
2015 13310
Name: arrival_date_year, dtype: int64
To count the number of guests who visited each year and

didnt cancelled the booking
In [98]: df[df['is_canceled']==0]['arrival_date_year'].value_counts().reset_index()
Out[98]:
index arrival_date_year
0 2016 31169
1 2017 21571
2 2015 10606
pie chart of the number of guests who visited each year
and didnt cancelled the booking
In [99]: data = df[df['is_canceled']==0]['arrival_date_year'].value_counts()
print(data)
labels = df['arrival_date_year'].unique()
explode = (0.0, 0.1, 0.1)
plt.pie(data, labels=labels, explode=explode, autopct='%0.2f')
plt.show()

2016 31169
2017 21571
2015 10606
Name: arrival_date_year, dtype: int64
To understand the number of visitors who havent

cancelled the bookings
In [100]: year_guest = df[df['is_canceled']==0]['arrival_date_year'].value_counts().reset_i
year_guest.columns = ['Year', 'Number of guest']
print(year_guest)

sns.lineplot(data=year_guest, x='Year', y='Number of guest')
plt.grid()
plt.show()
Year Number of guest
0 2016 31169
1 2017 21571
2 2015 10606
Out[101]: ['hotel',
'is_canceled',
'lead_time',
'adults',
'children',
'babies',
'meal',
'country',
'market_segment',
'booking_changes',
'deposit_type',
'agent',
'customer_type',
'adr',
To count number of children who visited the hotel and

didnt cancelled the booking
In [102]: children_guest = df[(df['is_canceled'] == 0) & (df['children'])]['hotel'].value_c
children_guest.columns = ['hotel', 'no of guests']
children_guest

Out[102]:
hotel no of guests
0 City Hotel 3233
1 Resort Hotel 2146

In [103]: children_guest = df[(df['is_canceled'] == 0) & (df['children'])]['hotel'].value_c
explode = (0.0, 0.05)
labels = df['hotel'].unique()
plt.figure(figsize=(12, 10))
plt.pie(children_guest, labels=labels, explode=explode, autopct='%0.2f')
plt.show()
To count number of adult who visited the hotel and didnt

cancelled the booking
In [104]: adult_guest = df[(df['is_canceled']==0) & (df['adults'])]['hotel'].value_counts()
adult_guest.columns = ['hotel', 'no of guests']
adult_guest
Out[104]:
hotel no of guests
0 City Hotel 11025
1 Resort Hotel 6159
In [105]: adult_guest = df[(df['is_canceled']==0) & (df['adults'])]['hotel'].value_counts()

explode = (0.0, 0.04)
plt.pie(adult_guest, labels=labels, explode=explode, autopct='%0.2f')
plt.show()
To count number of babies who visited the hotel and didnt

cancelled the booking
In [106]: babies_guest = df[(df['is_canceled']==0) & (df['babies'])]['hotel'].value_counts(
babies_guest.columns = ['hotel', 'no of guests']
babies_guest
Out[106]:
hotel no of guests
0 Resort Hotel 435
1 City Hotel 298
In [107]: babies_guest = df[(df['is_canceled']==0) & (df['babies'])]['hotel'].value_counts(

explode = (0.0, 0.04)
plt.pie(babies_guest, labels=labels, explode=explode, autopct='%0.2f')
plt.show()
visualisation of number of visitors (adults,children and

babies) visited by each country
In [108]: cancel_by_adults = df[(df['is_canceled']==0) & (df['adults'])]['hotel'].value_cou
cancel_by_adults.columns = ['hotel', 'no of guest']
# cancel_by_adults

cancel_by_children = df[(df['is_canceled']==0) & (df['children'])]['hotel'].value
cancel_by_children.columns = ['hotel', 'no of guest']
# cancel_by_children

cancel_by_babies = df[(df['is_canceled']==0) & (df['babies'])]['hotel'].value_cou
cancel_by_babies.columns = ['hotel', 'no of guest']
# cancel_by_adults

# visualize them all

sns.lineplot(data=cancel_by_adults, x=cancel_by_adults["hotel"],y=cancel_by_adult
sns.lineplot(data=cancel_by_children, x=cancel_by_children["hotel"],y=cancel_by_c
sns.lineplot(data=cancel_by_babies, x=cancel_by_babies["hotel"],y=cancel_by_babie

plt.show()

Booking cancel by each country

In [109]: country_guest = df[(df['is_canceled']==1)]['country'].value_counts().reset_index(
country_guest.columns=['country','no of guest']
country_guest

Out[109]:
country no of guest
0 PRT 9824
1 GBR 1985
2 ESP 1862
3 FRA 1733
4 ITA 1075
... ... ...
122 KHM 1
123 FJI 1
124 MCO 1
125 UMI 1
126 TJK 1
Booking not cancel by each country

In [110]: country_guest = df[(df['is_canceled']==0)]['country'].value_counts().reset_index(
country_guest.columns=['country','no of guest']
country_guest
Out[110]:
country no of guest
0 PRT 18058
1 GBR 8447
2 FRA 7104
3 ESP 5390
4 DEU 4334
... ... ...
160 MAC 1
161 CYM 1
162 TJK 1
163 ZMB 1
164 PLW 1
number of visitors visited by each country visual

representation
In [111]: # import plotly.express as px
# figure = px.line(country_guest, x='country', y='no of guest')
# figure.show()
Out[112]: ['hotel',
'is_canceled',
'lead_time',
'adults',
'children',
'babies',
'meal',
'country',
'market_segment',
'booking_changes',
'deposit_type',
'agent',
'customer_type',
'adr',
In [113]: count = df['assigned_room_type'].value_counts().reset_index()
count.columns = ['assignedroom', 'no of guest']
count

Out[113]:
assignedroom no of guest
0 A 46301
1 D 22422
2 E 7193
3 F 3625
4 G 2498
5 C 2165
6 B 1820
7 H 706
8 I 357
9 K 276
10 P 6
11 L 1
not canceled 'room type' by guests

In [114]: not_canceled = df[(df['is_canceled'])==0]['assigned_room_type'].value_counts().re
not_canceled.columns = ['assignedroom', 'no of guest']
not_canceled
Out[114]:
0 A 32120
1 D 16991
2 E 5495
3 F 2731
4 C 1770
5 G 1745
6 B 1421
7 H 457
8 I 352
9 K 264
canceled 'room type' by guests

In [115]: canceled = df[(df['is_canceled'])==1]['assigned_room_type'].value_counts().reset_
canceled.columns = ['assignedroom', 'no of guest']
canceled
Out[115]:
0 A 14181
1 D 5431
2 E 1698
3 F 894
4 G 753
5 B 399
6 C 395
7 H 249
8 K 12
9 P 6
10 I 5
11 L 1
Visulisation to represnt the above data

In [116]:
sns.lineplot(data=not_canceled, x=not_canceled["assignedroom"],y=not_canceled["no
sns.lineplot(data=canceled, x=canceled["assignedroom"],y=canceled["no of guest"])
plt.show()
representaation of assignedroom vs hotel

In [117]:
sns.countplot(data=df, x= df['assigned_room_type'], hue='hotel')
plt.show()

Out[118]: ['hotel',
'is_canceled',
'lead_time',
'adults',
'children',
'babies',
'meal',
'country',
'market_segment',
'booking_changes',
'deposit_type',
'agent',
'customer_type',
'adr',
Total Nights
In [119]: df['stays_in_weekend_nights']+df['stays_in_week_nights']
Out[119]: 0 0
1 0
2 1
3 1
4 2
..
119385 7
119386 7
119387 7
119388 7
119389 9
Length: 87370, dtype: int64
creating a column named 'total_nights' in dataset

In [120]: df['total_nights'] = df['stays_in_weekend_nights']+df['stays_in_week_nights']
In [121]: df.head()
Out[121]:
Resort
0 0 342 2015 July 27
Hotel
Resort
1 0 737 2015 July 27
Hotel
Resort
2 0 7 2015 July 27
Hotel
Resort
3 0 13 2015 July 27
Hotel
Resort
4 0 14 2015 July 27
Hotel
In [122]: df['total_nights']
Out[122]: 0 0
1 0
2 1
3 1
4 2
..
119385 7
119386 7
119387 7
119388 7
119389 9
Name: total_nights, Length: 87370, dtype: int64
In [123]: df.groupby(['hotel'])['total_nights'].count()
Out[123]: hotel
City Hotel 53426
Resort Hotel 33944
Name: total_nights, dtype: int64
Percentage of total_nights counts on hotels

In [124]: data = df.groupby(['hotel'])['total_nights'].count()
explode = (0.0, 0.1)
plt.pie(data, labels=labels, explode=explode, autopct="%0.2f")
plt.title("Percentage of total_nights counts on hotels")

plt.show()
In [125]: df.groupby(['hotel'])['total_nights'].value_counts()

Out[125]: hotel total_nights
City Hotel 3 13552
2 10824
1 10282
4 9620
5 4180
...
Resort Hotel 38 1
45 1
46 1
60 1
69 1
Name: total_nights, Length: 76, dtype: int64

EDA of Hotel Booking Dataset - Kaggle

Uploaded by

Copyright:

Available Formats

You might also like

EDA of Hotel Booking Dataset - Kaggle

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

EDA of Hotel Booking Dataset - Kaggle

Uploaded by

Copyright:

Available Formats

Mehedi Azad

Data Analyst | Data Science | Machine Learning |

In [ ]: #linkedin link: https://www.linkedin.com/in/mehediazad/

In [1]: import numpy as np

In [2]: df = pd.read_csv(r'G:\dataset\EDA ofhotel booking -kaggle\hotel_bookings.csv\hote

... ... ... ... ... ...

119390 rows × 32 columns

Checking First 5 records from this dataset using head()

Checking Last 5 records from this dataset using tail()

Checking columns of this dataset

Out[5]: Index(['hotel', 'is_canceled', 'lead_time', 'arrival_date_year',

'stays_in_week_nights', 'adults', 'children', 'babies', 'meal',

'country', 'market_segment', 'distribution_channel',

'assigned_room_type', 'booking_changes', 'deposit_type', 'agent',

'company', 'days_in_waiting_list', 'customer_type', 'adr',

Showing Statistical information regarding the data using

count 119390.000000 119390.000000 119390.000000 119390.000000 1193

mean 0.370416 104.011416 2016.156554 27.165173

std 0.482918 106.863097 0.707476 13.605138

min 0.000000 0.000000 2015.000000 1.000000

25% 0.000000 18.000000 2016.000000 16.000000

50% 0.000000 69.000000 2016.000000 28.000000

75% 1.000000 160.000000 2017.000000 38.000000

max 1.000000 737.000000 2017.000000 53.000000

Check the informations about the dataset using info()

RangeIndex: 119390 entries, 0 to 119389

Data columns (total 32 columns):

# Column Non-Null Count Dtype

--- ------ -------------- -----

0 hotel 119390 non-null object

1 is_canceled 119390 non-null int64

2 lead_time 119390 non-null int64

3 arrival_date_year 119390 non-null int64

4 arrival_date_month 119390 non-null object

5 arrival_date_week_number 119390 non-null int64

6 arrival_date_day_of_month 119390 non-null int64

7 stays_in_weekend_nights 119390 non-null int64

8 stays_in_week_nights 119390 non-null int64

9 adults 119390 non-null int64

10 children 119386 non-null float64

11 babies 119390 non-null int64

12 meal 119390 non-null object

13 country 118902 non-null object

14 market_segment 119390 non-null object

15 distribution_channel 119390 non-null object

16 is_repeated_guest 119390 non-null int64

17 previous_cancellations 119390 non-null int64

18 previous_bookings_not_canceled 119390 non-null int64

19 reserved_room_type 119390 non-null object

20 assigned_room_type 119390 non-null object

21 booking_changes 119390 non-null int64

22 deposit_type 119390 non-null object

23 agent 103050 non-null float64

24 company 6797 non-null float64

25 days_in_waiting_list 119390 non-null int64

26 customer_type 119390 non-null object

27 adr 119390 non-null float64