First of all I imported the libraries that are essential for our data analysis process.

We will do make our data

clean in order to perform analysis.

In [3]: import pandas as pd

import numpy as nm
import seaborn as sns
import matplotlib.pyplot as plt
import statsmodels.stats.multicomp as mc

import os
for dirname, _, filenames in os.walk('/kaggle/input'):
for filename in filenames:
print(os.path.join(dirname, filename))

I installed opendatastes to import data from the website of Kaggle.

In [4]: !pip install opendatasets

Requirement already satisfied: opendatasets in c:\users\47455\anaconda3\lib\site-package
s (0.1.22)
Requirement already satisfied: kaggle in c:\users\47455\anaconda3\lib\site-packages (fro
m opendatasets) (1.5.13)
Requirement already satisfied: tqdm in c:\users\47455\anaconda3\lib\site-packages (from
opendatasets) (4.64.0)
Requirement already satisfied: click in c:\users\47455\anaconda3\lib\site-packages (from
opendatasets) (8.0.4)
Requirement already satisfied: colorama in c:\users\47455\anaconda3\lib\site-packages (f
rom click->opendatasets) (0.4.4)
Requirement already satisfied: python-slugify in c:\users\47455\anaconda3\lib\site-packa
ges (from kaggle->opendatasets) (5.0.2)
Requirement already satisfied: urllib3 in c:\users\47455\anaconda3\lib\site-packages (fr
om kaggle->opendatasets) (1.26.9)
Requirement already satisfied: requests in c:\users\47455\anaconda3\lib\site-packages (f
rom kaggle->opendatasets) (2.27.1)
Requirement already satisfied: certifi in c:\users\47455\anaconda3\lib\site-packages (fr
om kaggle->opendatasets) (2021.10.8)
Requirement already satisfied: six>=1.10 in c:\users\47455\anaconda3\lib\site-packages
(from kaggle->opendatasets) (1.16.0)
Requirement already satisfied: python-dateutil in c:\users\47455\anaconda3\lib\site-pack
ages (from kaggle->opendatasets) (2.8.2)
Requirement already satisfied: text-unidecode>=1.3 in c:\users\47455\anaconda3\lib\site-
packages (from python-slugify->kaggle->opendatasets) (1.3)
Requirement already satisfied: idna<4,>=2.5 in c:\users\47455\anaconda3\lib\site-package
s (from requests->kaggle->opendatasets) (3.3)
Requirement already satisfied: charset-normalizer~=2.0.0 in c:\users\47455\anaconda3\lib
\site-packages (from requests->kaggle->opendatasets) (2.0.4)

In [5]: import opendatasets as od

In [6]: dataset = 'https://www.kaggle.com/datasets/arianazmoudeh/airbnbopendata'

In [7]: od.download(dataset)
Skipping, found downloaded files in ".\airbnbopendata" (use force=True to force downloa
d)

In [8]: import os

In [9]: data_dir = './airbnbopendata'

In [10]: os.listdir(data_dir)
 ['Airbnb_Open_Data.csv']
Out[10]:

In [11]: frame = pd.read_csv('Airbnb_Open_Data.csv',low_memory=False)

frame

Out[11]:
neighbourhood
id NAME host id host_identity_verified host name neighbourhood
group

Clean & quiet

0 1001254 apt home by 80014485718 unconfirmed Madaline Brooklyn Kensington
the park

Skylit Midtown
1 1002102 52335172823 verified Jenna Manhattan Midtown
Castle

THE VILLAGE
OF
2 1002403 78829239556 NaN Elise Manhattan Harlem
HARLEM....NEW
YORK !

3 1002755 NaN 85098326012 unconfirmed Garry Brooklyn Clinton Hill

Entire Apt:
Spacious
4 1003689 92037596077 verified Lyndon Manhattan East Harlem
Studio/Loft by
central park

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

Spare room in
102594 6092437 12312296767 verified Krik Brooklyn Williamsburg
Williamsburg

Best Location
Morningside
102595 6092990 near Columbia 77864383453 unconfirmed Mifan Manhattan
Heights
U

Comfy, bright
102596 6093542 room in 69050334417 unconfirmed Megan Brooklyn Park Slope
Brooklyn

Big Studio-One
102597 6094094 Stop from 11160591270 unconfirmed Christopher Queens Long Island City
Midtown

585 sf Luxury Upper West

102598 6094647 68170633372 unconfirmed Rebecca Manhattan
Studio Side

102599 rows × 26 columns

Removing the duplicates

In [12]: # Count the number of rows in the original DataFrame
print('Number of rows before removing duplicates:', len(frame))

# Drop duplicate rows based on all columns

frame = frame.drop_duplicates()

# Count the number of rows in the updated DataFrame

print('Number of rows after removing duplicates:', len(frame))

Number of rows before removing duplicates: 102599

Number of rows after removing duplicates: 102058

checking that do we have null values in our data.

In [13]: print(frame.isnull().sum())

id 0
NAME 250
host id 0
host_identity_verified 289
host name 404
neighbourhood group 29
neighbourhood 16
lat 8
long 8
country 532
country code 131
instant_bookable 105
cancellation_policy 76
room type 0
Construction year 214
price 247
service fee 273
minimum nights 400
number of reviews 183
last review 15832
reviews per month 15818
review rate number 319
calculated host listings count 319
availability 365 448
house_rules 51842
license 102056
dtype: int64

I wanted to see the column names so, I performed the below code

In [14]: print(frame.columns)
Index(['id', 'NAME', 'host id', 'host_identity_verified', 'host name',
'neighbourhood group', 'neighbourhood', 'lat', 'long', 'country',
'country code', 'instant_bookable', 'cancellation_policy', 'room type',
'Construction year', 'price', 'service fee', 'minimum nights',
'number of reviews', 'last review', 'reviews per month',
'review rate number', 'calculated host listings count',
'availability 365', 'house_rules', 'license'],
dtype='object')

In [15]: print(frame.head())
id NAME host id \
0 1001254 Clean & quiet apt home by the park 80014485718
1 1002102 Skylit Midtown Castle 52335172823
2 1002403 THE VILLAGE OF HARLEM....NEW YORK ! 78829239556
3 1002755 NaN 85098326012
4 1003689 Entire Apt: Spacious Studio/Loft by central park 92037596077
 host_identity_verified host name neighbourhood group neighbourhood \
0 unconfirmed Madaline Brooklyn Kensington
1 verified Jenna Manhattan Midtown
2 NaN Elise Manhattan Harlem
3 unconfirmed Garry Brooklyn Clinton Hill
4 verified Lyndon Manhattan East Harlem

lat long country ... service fee minimum nights \

0 40.64749 -73.97237 United States ... $193 10.0
1 40.75362 -73.98377 United States ... $28 30.0
2 40.80902 -73.94190 United States ... $124 3.0
3 40.68514 -73.95976 United States ... $74 30.0
4 40.79851 -73.94399 United States ... $41 10.0

number of reviews last review reviews per month review rate number \
0 9.0 10/19/2021 0.21 4.0
1 45.0 5/21/2022 0.38 4.0
2 0.0 NaN NaN 5.0
3 270.0 7/5/2019 4.64 4.0
4 9.0 11/19/2018 0.10 3.0

calculated host listings count availability 365 \

0 6.0 286.0
1 2.0 228.0
2 1.0 352.0
3 1.0 322.0
4 1.0 289.0

house_rules license
0 Clean up and treat the home the way you'd like... NaN
1 Pet friendly but please confirm with me if the... NaN
2 I encourage you to use my kitchen, cooking and... NaN
3 NaN NaN
4 Please no smoking in the house, porch or on th... NaN

[5 rows x 26 columns]

Checking the Shape of the data.

In [16]: print(frame.shape)
(102058, 26)

Used the mode function to fill the null values of each data.

In [17]: frame_copy=frame.copy()
frame_copy['long'].fillna(frame_copy['long'].mode()[0], inplace=True)
frame_copy['NAME'].fillna(frame_copy['NAME'].mode()[0], inplace=True)
frame_copy['neighbourhood group'].fillna(frame_copy['neighbourhood group'].mode()[0],inp
frame_copy['neighbourhood'].fillna(frame_copy['neighbourhood'].mode()[0],inplace = True)
frame_copy['host name'].fillna(frame_copy['host name'].mode()[0],inplace=True)
#replacing null values using mode in name column
frame_copy['lat'].fillna(frame_copy['lat'].mode()[0],inplace=True)
#replacing null values using mode in name column
frame_copy['long'].fillna(frame_copy['long'].mode()[0], inplace=True)
#replacing null values using mode in name column
frame_copy['country'].fillna(frame_copy['country'].mode()[0], inplace=True)
#replacing null values using mode in name column
frame_copy['country code'].fillna(frame_copy['country code'].mode()[0], inplace=True)
#replacing null values using mode in name column
frame_copy['instant_bookable'].fillna(frame_copy['instant_bookable'].mode()[0], inplace=
#replacing null values using mode in name column
frame_copy['cancellation_policy'].fillna(frame_copy['cancellation_policy'].mode()[0], inp
frame_copy['host_identity_verified'].fillna(frame_copy['host_identity_verified'].mode()[
 frame_copy['Construction year'].fillna(frame_copy['Construction year'].mode()[0], inplac
frame_copy['service fee'].fillna(frame_copy['service fee'].mode()[0], inplace=True)
frame_copy['minimum nights'].fillna(frame_copy['minimum nights'].mode()[0], inplace=True
frame_copy['number of reviews'].fillna(frame_copy['number of reviews'].mode()[0], inplac
frame_copy['last review'].fillna(frame_copy['last review'].mode()[0], inplace=True)
frame_copy['reviews per month'].fillna(frame_copy['reviews per month'].mode()[0], inplac
frame_copy['review rate number'].fillna(frame_copy['review rate number'].mode()[0], inpl
frame_copy['calculated host listings count'].fillna(frame_copy['calculated host listings
frame_copy['availability 365'].fillna(frame_copy['availability 365'].mode()[0], inplace=
frame_copy['house_rules'].fillna(frame_copy['house_rules'].mode()[0], inplace=True)
frame_copy['price'].fillna(frame_copy['price'].mode()[0], inplace=True)

In [18]: frame = frame_copy.copy()

In [19]: print(frame.isnull().sum())

id 0
NAME 0
host id 0
host_identity_verified 0
host name 0
neighbourhood group 0
neighbourhood 0
lat 0
long 0
country 0
country code 0
instant_bookable 0
cancellation_policy 0
room type 0
Construction year 0
price 0
service fee 0
minimum nights 0
number of reviews 0
last review 0
reviews per month 0
review rate number 0
calculated host listings count 0
availability 365 0
house_rules 0
license 102056
dtype: int64

to remove the license column I used the drop.

In [20]: frame = frame.drop("license", axis=1)

In [21]: print(frame.isnull().sum())

id 0
NAME 0
host id 0
host_identity_verified 0
host name 0
neighbourhood group 0
neighbourhood 0
lat 0
long 0
country 0
country code 0
instant_bookable 0
cancellation_policy 0
room type 0
 Construction year 0
price 0
service fee 0
minimum nights 0
number of reviews 0
last review 0
reviews per month 0
review rate number 0
calculated host listings count 0
availability 365 0
house_rules 0
dtype: int64

The 'types_of_rooms' variable contains the percentage of each type of room in the dataset, rounded to one
decimal point.

In [22]: types_of_rooms = frame['room type'].value_counts(normalize=True).mul(100).round(1)

types_of_rooms

Entire home/apt 52.4

Out[22]:
Private room 45.4
Shared room 2.2
Hotel room 0.1
Name: room type, dtype: float64

The explanation of the code below,

The first four lines of code convert the price column from object to float data type, by removing commas
and dollar signs, and then parsing it as numeric data using pandas. The next two lines of code create two
groups based on the room type (Entire home/apt and Private room) and extract the price values for each
group. The stats.ttest_ind() function from the scipy library is then used to perform an independent two-
sample t-test between the two groups, assuming unequal variances.

Finally, the t-statistic and p-value for the t-test are printed using f-strings.

The purpose of this code is to perform a statistical analysis to test whether there is a significant difference in
mean price between the two groups of listings (Entire home/apt and Private room). The t-test is a common
method for comparing means of two groups and determining whether the difference between them is
statistically significant or not.

In [23]: from scipy import stats

frame['price'] = frame['price'].astype(str)
frame['price'] = frame['price'].str.replace(',', '').str.replace('$', '', regex=False)
frame['price'] = frame['price'].astype(float)
frame['price'] = pd.to_numeric(frame['price'])
# Split data into two groups based on room type
group1 = frame[frame['room type'] == 'Entire home/apt']['price']
group2 = frame[frame['room type'] == 'Private room']['price']

# Perform t-test
t_stat, p_value = stats.ttest_ind(group1, group2, equal_var=False)

# Print results
print(f"T-statistic: {t_stat}")
print(f"P-value: {p_value}")

T-statistic: 0.10435722164673732
P-value: 0.9168860835285177

In the code below, I performed an ANOVA (Analysis of Variance) test to determine if there are significant
differences in mean price between three groups of listings, based on their neighbourhood_group feature.
In [24]: # create three groups based on neighbourhood_group
group1 = frame[frame['neighbourhood group'] == 'Brooklyn']['price']
group2 = frame[frame['neighbourhood group'] == 'Manhattan']['price']
group3 = frame[frame['neighbourhood group'] == 'Queens']['price']

# perform ANOVA test

f_statistic, p_value = stats.f_oneway(group1, group2, group3)

# print results
print("F-statistic:", f_statistic)
print("P-value:", p_value)

F-statistic: 3.1372044124800285
P-value: 0.043408308935618714

In [25]: # perform one-way ANOVA

f_stat, p_val = stats.f_oneway(frame[frame['neighbourhood group'] == 'Brooklyn']['price'
frame[frame['neighbourhood group'] == 'Manhattan']['price
frame[frame['neighbourhood group'] == 'Queens']['price'],
frame[frame['neighbourhood group'] == 'Staten Island']['p
frame[frame['neighbourhood group'] == 'Bronx']['price'])

# perform Tukey's HSD post-hoc test

m_comp = mc.MultiComparison(frame['price'], frame['neighbourhood group'])
tukey_res = m_comp.tukeyhsd()

print(tukey_res)

Multiple Comparison of Means - Tukey HSD, FWER=0.05

========================================================================
group1 group2 meandiff p-adj lower upper reject
------------------------------------------------------------------------
Bronx Brooklyn -1.1063 1.0 -20.5605 18.3478 False
Bronx Manhattan -5.0561 0.988 -24.4837 14.3715 False
Bronx Queens 2.4874 0.9998 -18.2014 23.1761 False
Bronx Staten Island -3.9998 0.9999 -40.9394 32.9399 False
Bronx brookln -46.6689 1.0 -1025.4278 932.09 False
Bronx manhatan -166.6689 0.9988 -1145.4278 812.09 False
Brooklyn Manhattan -3.9498 0.5913 -10.656 2.7565 False
Brooklyn Queens 3.5937 0.933 -6.1821 13.3695 False
Brooklyn Staten Island -2.8934 1.0 -35.0194 29.2325 False
Brooklyn brookln -45.5626 1.0 -1024.1516 933.0264 False
Brooklyn manhatan -165.5626 0.9989 -1144.1516 813.0264 False
Manhattan Queens 7.5435 0.2499 -2.1794 17.2663 False
Manhattan Staten Island 1.0563 1.0 -31.0536 33.1663 False
Manhattan brookln -41.6128 1.0 -1020.2013 936.9757 False
Manhattan manhatan -161.6128 0.999 -1140.2013 816.9757 False
Queens Staten Island -6.4871 0.9973 -39.3754 26.4012 False
Queens brookln -49.1562 1.0 -1027.7706 929.4581 False
Queens manhatan -169.1562 0.9987 -1147.7706 809.4581 False
Staten Island brookln -42.6691 1.0 -1021.7618 936.4236 False
Staten Island manhatan -162.6691 0.999 -1141.7618 816.4236 False
brookln manhatan -120.0 1.0 -1503.9172 1263.9172 False
------------------------------------------------------------------------

In [26]: print(frame['neighbourhood group'].unique())

['Brooklyn' 'Manhattan' 'brookln' 'manhatan' 'Queens' 'Staten Island'
'Bronx']

In [27]: frame['neighbourhood group'] = frame['neighbourhood group'].replace({'brookln': 'Brookly

In [28]: print(frame['neighbourhood group'].unique())

['Brooklyn' 'Manhattan' 'Queens' 'Staten Island' 'Bronx']

In [29]: print(frame['room type'].unique())

['Private room' 'Entire home/apt' 'Shared room' 'Hotel room']

Now, In the below code data analysis have been performed by using different types of graphs like barpchart,
piechart, boxplot, histogram, scatterplot and violinplot.

In [30]: # group data by room type and calculate average price

avg_price = frame.groupby('room type')['price'].mean()

# create barplot
fig, ax = plt.subplots()
ax.bar(avg_price.index, avg_price.values, color=['blue', 'green', 'red', 'orange'], widt

# add labels and title

ax.set_xlabel('Room Type')
ax.set_ylabel('Average Price')
ax.set_title('Barplot of Average Price by Room Type')

# create legend
colors = {'Private room': 'blue', 'Entire home/apt': 'green', 'Shared room': 'red', 'Hot
labels = list(colors.keys())
handles = [plt.Rectangle((0,0),1,1, color=colors[label]) for label in labels]
ax.legend(handles, labels, bbox_to_anchor=(1.05, 1), loc='upper left')

# display plot
plt.show()

In [31]: import matplotlib.pyplot as plt

# create a list of labels for each neighbourhood group

labels = ['Brooklyn', 'Manhattan', 'Queens', 'Staten Island', 'Bronx']

# create a list of colors for each neighbourhood group

colors = ['#1f77b4', '#ff7f0e', '#2ca02c', '#d62728', '#9467bd']

# filter the DataFrame to include only the neighbourhood group column

ng_counts = frame['neighbourhood group'].value_counts()

# create a pie chart of neighbourhood group distribution

plt.figure(figsize=(8,6))
plt.pie(ng_counts, labels=ng_counts.index, colors=colors, autopct='%1.1f%%')

# add legend outside the chart

plt.legend(labels=labels, bbox_to_anchor=(1.05, 1), loc='upper left', borderaxespad=0.)
 # set title
plt.title('Pie Chart of Neighbourhood Group Distribution', fontsize=14)

# show the plot

plt.show()

In [32]: # create a list of labels for each neighbourhood group

labels = ['Brooklyn', 'Manhattan', 'Queens', 'Staten Island', 'Bronx']

# define colors for each box

colors = ['#1f77b4', '#ff7f0e', '#2ca02c', '#d62728', '#9467bd']

# define box properties

boxprops = dict(linestyle='-', linewidth=2, color='black')

# create a boxplot of price by neighbourhood group

plt.figure(figsize=(8,6))
boxplots = plt.boxplot([frame[frame['neighbourhood group']=='Brooklyn']['price'],
frame[frame['neighbourhood group']=='Manhattan']['price'],
frame[frame['neighbourhood group']=='Queens']['price'],
frame[frame['neighbourhood group']=='Staten Island']['price'],
frame[frame['neighbourhood group']=='Bronx']['price']],
labels=labels, boxprops=boxprops, patch_artist=True)

# set colors for each box

for patch, color in zip(boxplots['boxes'], colors):
patch.set_facecolor(color)
patch.set_label('')

# set axis labels and title

plt.xlabel('Neighbourhood Group', fontsize=12)
plt.ylabel('Price', fontsize=12)
plt.title('Boxplot of Price by Neighbourhood Group', fontsize=14)

# add legend
plt.legend(handles=boxplots['boxes'], labels=labels, bbox_to_anchor=(1.05, 1), loc='uppe

# show the plot

plt.show()
In [33]: # Create a horizontal bar chart
colors = ['#5DA5DB', '#FAA43C', '#60BD69', '#F17CB1']
plt.figure(figsize=(8, 6))
ax = plt.gca()
ax.barh(types_of_rooms.index, types_of_rooms.values, color=colors)
plt.title('Percentage of Types of Rooms')
plt.xlabel('Percentage')
plt.show()

In [34]: plt.figure(figsize=(8, 6))

sns.histplot(data=frame, x='price', kde=True,color='orange')
plt.title('Histogram of Price')
plt.xlabel('Price')
plt.ylabel('Frequency')
plt.show()
In [35]: import matplotlib.pyplot as plt
color = ['purple']
plt.figure(figsize=(8, 6))
plt.scatter(frame['price'], frame['number of reviews'],color=color)
plt.title('Scatter Plot of Price vs. Number of Reviews')
plt.xlabel('price')
plt.ylabel('number of reviews')
plt.show()

In [37]: sns.violinplot(x='room type', y='reviews per month', data=frame, palette='Blues')

plt.title('Distribution of Reviews per Month by Room Type')
plt.show()