Scribd
Upload
12 views10 pages

Coding

The document outlines a data science project focused on hotel booking analysis using Python libraries such as NumPy, Pandas, and SciPy. It includes steps for data manipulation, cleaning, and statistical analysis, along with linear and logistic regression modeling. Key findings include descriptive statistics and the performance of regression models on hotel booking data.

Uploaded by

rsbcpkqd9d
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
12 views10 pages

Coding

The document outlines a data science project focused on hotel booking analysis using Python libraries such as NumPy, Pandas, and SciPy. It includes steps for data manipulation, cleaning, and statistical analysis, along with linear and logistic regression modeling. Key findings include descriptive statistics and the performance of regression models on hotel booking data.

Uploaded by

rsbcpkqd9d
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

4/19/26, 11:55 PM Data_Science_Project_Final (1).

ipynb - Colab

keyboard_arrow_down Hotel Booking Data Science Analysis

keyboard_arrow_down Install & Explore NumPy, Pandas, SciPy, Seaborn

import numpy as np
import pandas as pd
import [Link] as plt
import seaborn as sns
import [Link] as stats
import os, kagglehub
from [Link] import norm
from mpl_toolkits.mplot3d import Axes3D
from sklearn.linear_model import LinearRegression, LogisticRegression
from sklearn.model_selection import train_test_split
from sklearn import metrics
from [Link] import MinMaxScaler

sns.set_theme(style="whitegrid")
print(f"NumPy {np.__version__} | Pandas {pd.__version__} | Seaborn {sns.__version__}")

NumPy 2.0.2 | Pandas 2.2.2 | Seaborn 0.13.2

keyboard_arrow_down Working with NumPy Arrays

# NumPy Array Operations


n = [Link]([1, 2, 3, 5, 7, 2, 4, 6, 3, 1])
r = [Link](50, 201, 15)

print(f"Max: {[Link]()} | Min: {[Link]()} | Mean: {[Link]():.2f}")


print("Reshaped (2x5):\n", [Link](2, 5))
print("Square Root:", [Link]([Link](n), 2))
print("Logarithm:", [Link]([Link](r), 2))

# Random and Linear space


[Link](42)
print("Zeros:", [Link](5))
print("Ones:", [Link](5))
print("Random Normal:", [Link](5).round(3))
print("Linspace:", [Link]([Link](0, 1, 12), 2))

# Matrix Math
g = [Link](9).reshape(3, 3)
print("Matrix G:\n", g)
print("G + G:\n", g + g)
print("G * G:\n", g * g)

Max: 7 | Min: 1 | Mean: 3.40


Reshaped (2x5):
[[1 2 3 5 7]
[2 4 6 3 1]]
Square Root: [1. 1.41 1.73 2.24 2.65 1.41 2. 2.45 1.73 1. ]
Logarithm: [3.91 4.17 4.38 4.55 4.7 4.83 4.94 5.04 5.14 5.22 5.3 ]
Zeros: [0. 0. 0. 0. 0.]
Ones: [1. 1. 1. 1. 1.]
Random Normal: [ 0.497 -0.138 0.648 1.523 -0.234]
Linspace: [0. 0.09 0.18 0.27 0.36 0.45 0.55 0.64 0.73 0.82 0.91 1. ]
Matrix G:
[[0 1 2]
[3 4 5]
[6 7 8]]
G + G:
[[ 0 2 4]
[ 6 8 10]
[12 14 16]]
G * G:
[[ 0 1 4]
[ 9 16 25]
[36 49 64]]

[Link] 1/10
4/19/26, 11:55 PM Data_Science_Project_Final (1).ipynb - Colab

keyboard_arrow_down Working with Pandas DataFrames

# Pandas DataFrames Basics


data = {
'hotel_type': ['City', 'Resort', 'City', 'Resort', 'City'],
'nights': [3, 7, 2, 5, 1],
'adr': [120.5, 85, 200, 95, 150],
'canceled': [0, 0, 1, 0, 1]
}
df_sample = [Link](data)
df_sample['revenue'] = df_sample['nights'] * df_sample['adr']

print("Sample DataFrame:\n", df_sample)


print("\nCanceled Bookings:\n", df_sample[df_sample['canceled'] == 1])
print("\nSlicing (Rows 1-3, nights & adr):\n", df_sample.loc[1:3, ['nights', 'adr']])

# Multi-Index Example
index = [Link].from_tuples([('Q1', 'Jan'), ('Q1', 'Feb'), ('Q2', 'Jan'), ('Q2', 'Feb')])
dm = [Link]([Link](4, 2), index=index, columns=['City', 'Resort'])
print("\nMulti-Index DataFrame:\n", dm)
print("\nAccessing Q1:\n", [Link]['Q1'])

Sample DataFrame:
hotel_type nights adr canceled revenue
0 City 3 120.5 0 361.5
1 Resort 7 85.0 0 595.0
2 City 2 200.0 1 400.0
3 Resort 5 95.0 0 475.0
4 City 1 150.0 1 150.0

Canceled Bookings:
hotel_type nights adr canceled revenue
2 City 2 200.0 1 400.0
4 City 1 150.0 1 150.0

Slicing (Rows 1-3, nights & adr):


nights adr
1 7 85.0
2 2 200.0
3 5 95.0

Multi-Index DataFrame:
City Resort
Q1 Jan -0.234137 1.579213
Feb 0.767435 -0.469474
Q2 Jan 0.542560 -0.463418
Feb -0.465730 0.241962

Accessing Q1:
City Resort
Jan -0.234137 1.579213
Feb 0.767435 -0.469474

keyboard_arrow_down Reading Data & Descriptive Analytics

path = kagglehub.dataset_download("mojtaba142/hotel-booking")
csv_path = next([Link](r, f) for r, d, fs in [Link](path) for f in fs if [Link]('.csv'))
df = pd.read_csv(csv_path)
display([Link]())
[Link]()
print("\nMissing:\n", [Link]().sum().nlargest(5))

[Link] 2/10
4/19/26, 11:55 PM Data_Science_Project_Final (1).ipynb - Colab

Using Colab cache for faster access to the 'hotel-booking' dataset.

hotel is_canceled lead_time arrival_date_year arrival_date_month arrival_date_week_number arrival_date_day_of_month st

Resort
0 0 342 2015 July 27 1
Hotel

Resort
1 0 737 2015 July 27 1
Hotel

Resort
2 0 7 2015 July 27 1
Hotel

Resort
3 0 13 2015 July 27 1
Hotel

Resort
4 0 14 2015 July 27 1
Hotel

5 rows × 36 columns
<class '[Link]'>
RangeIndex: 119390 entries, 0 to 119389
Data columns (total 36 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
32 name 119390 non-null object
33 email 119390 non-null object
34 phone-number 119390 non-null object
35 credit_card 119390 non-null object
dtypes: float64(4), int64(16), object(16)
memory usage: 32.8+ MB

Missing:
company 112593
agent 16340
country 488
children 4
hotel 0
dtype: int64

df = [Link]({'children': 0, 'agent': 0, 'company': 0, 'country': 'Unknown'}).drop_duplicates()


df = df[(df[['adults', 'children', 'babies']].sum(axis=1) > 0) & df['adr'].between(0, 5000)]
df = [Link](columns={'adr': 'avg_daily_rate'})
print(f"Cleaned Shape: {[Link]}")

[Link] 3/10
4/19/26, 11:55 PM Data_Science_Project_Final (1).ipynb - Colab
for col in ['hotel', 'country', 'meal']:
print(f"\nTop 5 {col}:\n", df[col].value_counts().head(5))

Cleaned Shape: (119208, 36)

Top 5 hotel:
hotel
City Hotel 79162
Resort Hotel 40046
Name: count, dtype: int64

Top 5 country:
country
PRT 48482
GBR 12119
FRA 10401
ESP 8560
DEU 7285
Name: count, dtype: int64

Top 5 meal:
meal
BB 92234
HB 14458
SC 10549
Undefined 1169
FB 798
Name: count, dtype: int64

keyboard_arrow_down Statistical Analysis

Univariate Analysis

Computing mean, median, mode, variance, std deviation, and skewness.

for col in ['avg_daily_rate', 'stays_in_week_nights']:


s = df[col]
print(f"\n--- {col} ---\nMean: {[Link]():.2f} | Median: {[Link]():.2f} | Std: {[Link]():.2f} | Skew: {[Link](s):.3f}")
print(f"Percentiles: {[Link](s, [25, 50, 75, 90])}")

--- avg_daily_rate ---


Mean: 101.93 | Median: 94.95 | Std: 48.04 | Skew: 1.030
Percentiles: [ 69.5 94.95 126. 164.05]

--- stays_in_week_nights ---


Mean: 2.50 | Median: 2.00 | Std: 1.90 | Skew: 2.755
Percentiles: [1. 2. 3. 5.]

keyboard_arrow_down Bivariate Analysis – Linear & Logistic Regression

def fit_eval(X, y, kind='linear'):


xt, xv, yt, yv = train_test_split(X, y, test_size=0.2, random_state=42)
model = LinearRegression().fit(xt, yt) if kind=='linear' else LogisticRegression(max_iter=1000).fit(xt, yt)
pred = [Link](xv)
if kind=='linear':
print(f"[Linear] R2: {metrics.r2_score(yv, pred):.4f} | MAE: {metrics.mean_absolute_error(yv, pred):.2f}")
else:
print(f"[Logistic] Accuracy: {metrics.accuracy_score(yv, pred):.4f}\n", metrics.classification_report(yv, pred))
return model, xv, yv

lin_m, x_l, y_l = fit_eval(df[['lead_time']], df['avg_daily_rate'])


log_feats = ['lead_time', 'stays_in_week_nights', 'adults', 'avg_daily_rate']
log_m, _, _ = fit_eval(df[log_feats], df['is_canceled'], 'logistic')

[Link](figsize=(10, 5))
sidx = x_l['lead_time'].argsort()
[Link](x_l, y_l, alpha=0.1, s=5)
[Link](x_l.iloc[sidx], lin_m.predict(x_l.iloc[sidx]), 'r', lw=2)
[Link]('Lead Time vs ADR'); [Link]()

[Link] 4/10
4/19/26, 11:55 PM Data_Science_Project_Final (1).ipynb - Colab

[Linear] R2: 0.0043 | MAE: 36.12


[Logistic] Accuracy: 0.6624
precision recall f1-score support

0 0.68 0.88 0.77 14909


1 0.60 0.29 0.39 8933

accuracy 0.66 23842


macro avg 0.64 0.59 0.58 23842
weighted avg 0.65 0.66 0.63 23842

keyboard_arrow_down Multiple Regression Analysis

mf = ['lead_time', 'stays_in_week_nights', 'stays_in_weekend_nights', 'adults', 'children', 'total_of_special_requests']


mm = LinearRegression().fit(df[mf], df['avg_daily_rate'])
print("Coefficients:", dict(zip(mf, mm.coef_.round(4))))
display(df[mf + ['avg_daily_rate']].corr().round(3))

Coefficients: {'lead_time': np.float64(-0.0343), 'stays_in_week_nights': np.float64(1.0762), 'stays_in_weekend_nights': [Link]


lead_time stays_in_week_nights stays_in_weekend_nights adults children total_of_special_requests a

lead_time 1.000 0.167 0.086 0.118 -0.038 -0.096

stays_in_week_nights 0.167 1.000 0.494 0.096 0.045 0.069

stays_in_weekend_nights 0.086 0.494 1.000 0.095 0.046 0.073

adults 0.118 0.096 0.095 1.000 0.029 0.123

children -0.038 0.045 0.046 0.029 1.000 0.082

total_of_special_requests -0.096 0.069 0.073 0.123 0.082 1.000

avg_daily_rate -0.068 0.071 0.054 0.235 0.341 0.182

keyboard_arrow_down Plotting Functions

Normal Curves

fig, axes = [Link](1, 2, figsize=(14, 4))


for ax, (col, c) in zip(axes, [('avg_daily_rate', 'royalblue'), ('lead_time', 'tomato')]):
x = [Link](df[col].min(), df[col].max(), 100)
[Link](x, [Link](x, df[col].mean(), df[col].std()), color=c, lw=2)
ax.set_title(f'Normal: {col}')
[Link]()

[Link] 5/10
4/19/26, 11:55 PM Data_Science_Project_Final (1).ipynb - Colab

keyboard_arrow_down Density and Contour Plots

fig, axes = [Link](1, 3, figsize=(18, 5))


[Link]('hotel')['avg_daily_rate'].[Link](ax=axes[0], legend=True, title='ADR Density')
smp = [Link](2000, random_state=42)
axes[1].scatter(smp.lead_time, smp.avg_daily_rate, alpha=0.1, s=5)
h2, xe, ye = np.histogram2d(smp.lead_time, smp.avg_daily_rate, bins=30)
xc, yc = 0.5*(xe[:-1]+xe[1:]), 0.5*(ye[:-1]+ye[1:])
axes[2].contourf(xc, yc, h2.T, cmap='RdYlGn')
[Link]()

# Correlation Heatmap and Color-coded Scatter


analysis_cols = ['lead_time', 'stays_in_week_nights', 'stays_in_weekend_nights', 'adults', 'avg_daily_rate', 'total_of_special_re
fig, (ax_heat, ax_scat) = [Link](1, 2, figsize=(16, 6))

# Heatmap
[Link](df[analysis_cols].corr(), annot=True, fmt='.2f', cmap='coolwarm', ax=ax_heat)
ax_heat.set_title('Correlation Heatmap')

# Scatter with Cancellation mapping


cancel_colors = df['is_canceled'].map({0: 'green', 1: 'red'})
ax_scat.scatter(df['adults'], df['avg_daily_rate'], c=cancel_colors, alpha=0.1, s=5)
ax_scat.set(title='ADR vs Adults (Green=Success, Red=Canceled)', xlabel='Adults', ylabel='ADR')

plt.tight_layout()
[Link]()

[Link] 6/10
4/19/26, 11:55 PM Data_Science_Project_Final (1).ipynb - Colab

keyboard_arrow_down Histograms

months = ['January', 'February', 'March', 'April', 'May', 'June', 'July', 'August', 'September', 'October', 'November', 'Decemb
df['arrival_date_month'] = [Link](df['arrival_date_month'], categories=months, ordered=True)
fig, axes = [Link](2, 3, figsize=(15, 8))
for i, (col, c) in enumerate([('avg_daily_rate','b'), ('lead_time','r'), ('stays_in_week_nights','g')]):
axes[0,i].hist(df[col], bins=40, color=c)
[Link].value_counts().[Link](ax=axes[1,0], rot=0)
df.arrival_date_month.value_counts().sort_index().[Link](ax=axes[1,1])
axes[1,2].hist2d(df.lead_time, df.avg_daily_rate, bins=30, cmap='Blues')
plt.tight_layout(); [Link]()

keyboard_arrow_down Three-Dimensional Plotting

s3 = [Link](1000, random_state=42)
fig = [Link](figsize=(15, 10))
ax1 = fig.add_subplot(221, projection='3d')
[Link](s3.lead_time, s3.stays_in_week_nights, s3.avg_daily_rate, c=s3.avg_daily_rate)
xs2 = [Link](-3, 3, 50); X2, Y2 = [Link](xs2, xs2); Z2 = [Link]([Link](X2**2 + Y2**2))
fig.add_subplot(222, projection='3d').plot_surface(X2, Y2, Z2, cmap='coolwarm')
fig.add_subplot(223, projection='3d').plot_wireframe(X2, Y2, Z2)
ax4 = fig.add_subplot(224, projection='3d')
ax4.bar3d(s3.lead_time[:20], s3.avg_daily_rate[:20], 0, 5, 5, s3.stays_in_week_nights[:20])
[Link]()

[Link] 7/10
4/19/26, 11:55 PM Data_Science_Project_Final (1).ipynb - Colab

keyboard_arrow_down Geographic Data Visualization with Basemap

Guest countries are plotted on a world map using country-level data.

cc = [Link].value_counts().head(10)
fig, (ax1, ax2) = [Link](1, 2, figsize=(15, 5))
[Link](ax=ax1); [Link](cc, labels=[Link], autopct='%1.1f%%'); [Link]()

[Link] 8/10
4/19/26, 11:55 PM Data_Science_Project_Final (1).ipynb - Colab

keyboard_arrow_down Full EDA Dashboard

Combined visualisation covering all important business insights.

fig, axes = [Link](3, 3, figsize=(18, 14))


([Link]('hotel')['is_canceled'].mean()*100).[Link](ax=axes[0,0], rot=0, title='Cancel %')
[Link](x='hotel', y='avg_daily_rate', data=df, ax=axes[0,1]).set(ylim=(0, 400))
df.arrival_date_month.value_counts().sort_index().plot(ax=axes[0,2], marker='o')
for i, col in enumerate(['customer_type', 'meal', 'distribution_channel']):
df[col].value_counts().[Link](ax=axes[1,i], title=col)
cnt, edges = [Link](df.avg_daily_rate, bins=50, density=True)
pdf = cnt/[Link]()
axes[2,0].plot(edges[1:], pdf, label='PDF'); axes[2,0].plot(edges[1:], [Link](pdf), label='CDF'); axes[2,0].legend()
[Link](df[['lead_time','avg_daily_rate','is_canceled']].corr(), annot=True, ax=axes[2,1])
axes[2,2].hexbin(df.lead_time, df.avg_daily_rate, gridsize=20, cmap='Blues')
plt.tight_layout(); [Link]()

keyboard_arrow_down Cancellation Risk Tier Calculator

df['risk_score'] = MinMaxScaler(feature_range=(0,100)).fit_transform(log_m.predict_proba(df[log_feats])[:,1].reshape(-1,1)).rou
df['risk_tier'] = [Link](df['risk_score'], bins=[0, 35, 65, 100], labels=['Low', 'Medium', 'High'], include_lowest=True)
df['deposit'] = df['avg_daily_rate'] * df['risk_tier'].map({'Low':0.1, 'Medium':0.25, 'High':0.5}).astype(float)
df['rev'] = df['avg_daily_rate'] * (df['stays_in_week_nights'] + df['stays_in_weekend_nights'])

display([Link]('risk_tier', observed=True).agg(bookings=('risk_score','count'), cancel_rate=('is_canceled','mean'), avg_dep

[Link] 9/10
4/19/26, 11:55 PM Data_Science_Project_Final (1).ipynb - Colab
fig, axes = [Link](1, 3, figsize=(18, 5))
clrs = {'Low':'seagreen', 'Medium':'goldenrod', 'High':'tomato'}
for t, c in [Link](): axes[0].hist(df[df['risk_tier']==t]['risk_score'], bins=20, alpha=0.6, color=c, label=t)
([Link]('risk_tier', observed=True)['is_canceled'].mean()*100).plot(kind='bar', ax=axes[1], color=[Link]())

canc = df[df['is_canceled']==1]
l, r = [Link]('risk_tier', observed=True)['rev'].sum()/1e6, [Link]('risk_tier', observed=True)['deposit'].sum()/1e6
axes[2].bar([Link](), l, color='grey', alpha=0.3); axes[2].bar([Link](), r, color=[Link]())
[Link]()
print(f"Lost: ${[Link]():,.2f} | Rec: ${[Link]():,.2f} | Rate: {([Link]()/[Link]())*100:.2f}%")

bookings cancel_rate avg_dep_pct

risk_tier

Low 76868 0.286 0.10

Medium 35083 0.495 0.25

High 7257 0.670 0.50

$ | $ |

[Link] 10/10

You might also like