Scribd
Upload
63 views9 pages

Airbnb Case - SHIRIN MAHAJAN Python

This document discusses analyzing airbnb data from Amsterdam. It loads and inspects an Airbnb case CSV file with over 18,000 rows and 20 columns of data on listings. It is found that the most common room type is 'Entire home/apt' and there are 14978 of these. When filtering the data to only shared rooms, 3376 listings have a price below 100. The 63 rows corresponding to shared room types are then dropped from the dataset since this count is very low compared to other room types.

Uploaded by

Shirin Mahajan
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
63 views9 pages

Airbnb Case - SHIRIN MAHAJAN Python

This document discusses analyzing airbnb data from Amsterdam. It loads and inspects an Airbnb case CSV file with over 18,000 rows and 20 columns of data on listings. It is found that the most common room type is 'Entire home/apt' and there are 14978 of these. When filtering the data to only shared rooms, 3376 listings have a price below 100. The 63 rows corresponding to shared room types are then dropped from the dataset since this count is very low compared to other room types.

Uploaded by

Shirin Mahajan
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
You are on page 1/ 9

9/9/21, 10:35 AM Airbnb Case _SHIRIN MAHAJAN

In [1]:
#1 Read the CSV file and create pandas dataframe object airbnb.

import numpy as np

import pandas as pd

bnb=pd.read_csv('Airbnbcase.csv')
bnb

Out[1]: room_id survey_id host_id room_type country city borough neighborhood

Shared De Pijp /
0 10176931 1476 49180562 NaN Amsterdam NaN
room Rivierenbuurt

Shared
1 8935871 1476 46718394 NaN Amsterdam NaN Centrum West
room

Shared
2 14011697 1476 10346595 NaN Amsterdam NaN Watergraafsmeer
room

Shared
3 6137978 1476 8685430 NaN Amsterdam NaN Centrum West
room

Shared De Baarsjes /
4 18630616 1476 70191803 NaN Amsterdam NaN
room Oud West

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

Private
18718 17789893 1476 47501089 NaN Amsterdam NaN Bijlmer Centrum
room

Private
18719 16877166 1476 67093870 NaN Amsterdam NaN Bijlmer Centrum
room

Private Geuzenveld /
18720 19859427 1476 29724632 NaN Amsterdam NaN
room Slotermeer

Private
18721 17132164 1476 115156569 NaN Amsterdam NaN Centrum West
room

Private
18722 7605782 1476 39503013 NaN Amsterdam NaN Centrum West
room

18723 rows × 20 columns

In [2]:
#2 Review first 5 rows

bnb.head(5)

localhost:8888/nbconvert/html/Airbnb Case _SHIRIN MAHAJAN.ipynb?download=false 1/9


9/9/21, 10:35 AM Airbnb Case _SHIRIN MAHAJAN

Out[2]: room_id survey_id host_id room_type country city borough neighborhood revie

Shared De Pijp /
0 10176931 1476 49180562 NaN Amsterdam NaN
room Rivierenbuurt

Shared
1 8935871 1476 46718394 NaN Amsterdam NaN Centrum West
room

Shared
2 14011697 1476 10346595 NaN Amsterdam NaN Watergraafsmeer
room

Shared
3 6137978 1476 8685430 NaN Amsterdam NaN Centrum West
room

Shared De Baarsjes /
4 18630616 1476 70191803 NaN Amsterdam NaN
room Oud West

In [3]:
#3 What are the room_type available

bnb['room_type'].value_counts()

Out[3]: Entire home/apt 14978

Private room 3682

Shared room 63

Name: room_type, dtype: int64

In [4]:
#4 How many rows and columns are present in the dataset

rows=len(bnb.axes[0])

cols=len(bnb.axes[1])

print('no. of rows:',rows)

print('no. of col:',cols)

no. of rows: 18723

no. of col: 20

In [5]:
#5 Count the number of locations where the price is below 100 in shared rooms

bnb['low_price'] = np.where(bnb['price'] < 100 , True, False)


bnb.head(5)

Out[5]: room_id survey_id host_id room_type country city borough neighborhood revie

Shared De Pijp /
0 10176931 1476 49180562 NaN Amsterdam NaN
room Rivierenbuurt

Shared
1 8935871 1476 46718394 NaN Amsterdam NaN Centrum West
room

Shared
2 14011697 1476 10346595 NaN Amsterdam NaN Watergraafsmeer
room

localhost:8888/nbconvert/html/Airbnb Case _SHIRIN MAHAJAN.ipynb?download=false 2/9


9/9/21, 10:35 AM Airbnb Case _SHIRIN MAHAJAN

room_id survey_id host_id room_type country city borough neighborhood revie

Shared
3 6137978 1476 8685430 NaN Amsterdam NaN Centrum West
room

Shared De Baarsjes /
4 18630616 1476 70191803 NaN Amsterdam NaN
room Oud West

5 rows × 21 columns

In [6]:
(bnb['low_price']==True).sum()

Out[6]: 3376

In [7]:
#6 Drop the shared room type as its count is very low compare to others

bnb.drop(bnb.loc[bnb['room_type']== 'Shared room'].index)

bnb.tail(5)

Out[7]: room_id survey_id host_id room_type country city borough neighborhood

Private Bijlmer
18718 17789893 1476 47501089 NaN Amsterdam NaN
room Centrum

Private Bijlmer
18719 16877166 1476 67093870 NaN Amsterdam NaN
room Centrum

Private Geuzenveld /
18720 19859427 1476 29724632 NaN Amsterdam NaN
room Slotermeer

Private
18721 17132164 1476 115156569 NaN Amsterdam NaN Centrum West
room

Private
18722 7605782 1476 39503013 NaN Amsterdam NaN Centrum West
room

5 rows × 21 columns

In [8]:
(bnb['room_type']=='Private room').sum()

Out[8]: 3682

In [9]:
(bnb['room_type']=='Entire home/apt').sum()

localhost:8888/nbconvert/html/Airbnb Case _SHIRIN MAHAJAN.ipynb?download=false 3/9


9/9/21, 10:35 AM Airbnb Case _SHIRIN MAHAJAN

Out[9]: 14978

In [10]:
#7Take a count of each neighborhood to show how popular they are w.r.t to airbnb apa
bnb['neighborhood'].value_counts()

Out[10]: De Baarsjes / Oud West 3289

De Pijp / Rivierenbuurt 2378

Centrum West 2225

Centrum Oost 1730

Westerpark 1430

Noord-West / Noord-Midden 1418

Oud Oost 1169

Bos en Lommer 988

Oostelijk Havengebied / Indische Buurt 921

Watergraafsmeer 517

Oud Noord 494

Ijburg / Eiland Zeeburg 378

Slotervaart 349

Buitenveldert / Zuidas 250

Noord West 241

Noord Oost 221

Geuzenveld / Slotermeer 195

Osdorp 163

De Aker / Nieuw Sloten 114

Bijlmer Centrum 99

Bijlmer Oost 97

Gaasperdam / Driemond 42

Westpoort 15

Name: neighborhood, dtype: int64

In [11]:
#8 What is the overall_satisfication count by each rating

os = bnb.groupby('overall_satisfaction', dropna=False).reviews.count()

os

Out[11]: overall_satisfaction

0.0 5748

1.0 1

1.5 1

2.5 1

3.0 19

3.5 109

4.0 577

4.5 4559

5.0 7708

Name: reviews, dtype: int64

In [25]:
bnb.drop(['pricevalue','low_price',],axis=1,inplace=True)

In [ ]:
bnb.drop(['low price',],axis=1,inplace=True)

In [13]:
#9 Create a sub dataset where overall satisfication is 0

rating = bnb[bnb.overall_satisfaction == 0]

rating.sample(10)

Out[13]: room_id survey_id host_id room_type country city borough neighborhood

Entire
10940 3172092 1476 16083490 NaN Amsterdam NaN Slotervaart
home/apt

localhost:8888/nbconvert/html/Airbnb Case _SHIRIN MAHAJAN.ipynb?download=false 4/9


9/9/21, 10:35 AM Airbnb Case _SHIRIN MAHAJAN

room_id survey_id host_id room_type country city borough neighborhood

Entire
9836 12983144 1476 71484940 NaN Amsterdam NaN Watergraafsmeer
home/apt

Entire
7350 17819035 1476 8986411 NaN Amsterdam NaN Westerpark
home/apt

Entire De Pijp /
9202 19888201 1476 10366758 NaN Amsterdam NaN
home/apt Rivierenbuurt

Entire Buitenveldert /
1709 14366189 1476 46858505 NaN Amsterdam NaN
home/apt Zuidas

Entire
6544 15780441 1476 99163836 NaN Amsterdam NaN Oud Oost
home/apt

Entire Noord-West /
5272 18559464 1476 8874273 NaN Amsterdam NaN
home/apt Noord-Midden

Entire
12714 14553702 1476 38953420 NaN Amsterdam NaN Westerpark
home/apt

Entire De Baarsjes /
11472 14165297 1476 85697893 NaN Amsterdam NaN
home/apt Oud West

Entire
4347 6877495 1476 36039260 NaN Amsterdam NaN Centrum West
home/apt

10 rows × 21 columns

In [14]:
#10 What is the satisfication ratings count where bedroom is 3

a= bnb[bnb.bedrooms == 3].bedrooms.count()

b = bnb[bnb.bedrooms == 3].overall_satisfaction.count()

print('Total apartments with 3 bedrooms:',a)

print('with ratings:',b)

Total apartments with 3 bedrooms: 1444

with ratings: 1444

localhost:8888/nbconvert/html/Airbnb Case _SHIRIN MAHAJAN.ipynb?download=false 5/9


9/9/21, 10:35 AM Airbnb Case _SHIRIN MAHAJAN

In [16]:
#11 On which price most of the properties are available

pl = bnb.groupby('price', dropna=False).room_id.count()

pl.sort_values(ascending = False)

Out[16]: price

119.0 1023

180.0 1001

144.0 887

150.0 621

132.0 588

...

373.0 1

371.0 1

369.0 1

367.0 1

6000.0 1

Name: room_id, Length: 423, dtype: int64

In [17]:
#12 Drop the values where bedroom are equal to zero

bnb = bnb.drop(bnb[bnb.bedrooms== 0].index)

bnb

Out[17]: room_id survey_id host_id room_type country city borough neighborhood

Shared De Pijp /
0 10176931 1476 49180562 NaN Amsterdam NaN
room Rivierenbuurt

Shared
1 8935871 1476 46718394 NaN Amsterdam NaN Centrum West
room

Shared
2 14011697 1476 10346595 NaN Amsterdam NaN Watergraafsmeer
room

Shared
3 6137978 1476 8685430 NaN Amsterdam NaN Centrum West
room

Shared De Baarsjes /
4 18630616 1476 70191803 NaN Amsterdam NaN
room Oud West

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

Private
18718 17789893 1476 47501089 NaN Amsterdam NaN Bijlmer Centrum
room

Private
18719 16877166 1476 67093870 NaN Amsterdam NaN Bijlmer Centrum
room

Private Geuzenveld /
18720 19859427 1476 29724632 NaN Amsterdam NaN
room Slotermeer

localhost:8888/nbconvert/html/Airbnb Case _SHIRIN MAHAJAN.ipynb?download=false 6/9


9/9/21, 10:35 AM Airbnb Case _SHIRIN MAHAJAN

room_id survey_id host_id room_type country city borough neighborhood

Private
18721 17132164 1476 115156569 NaN Amsterdam NaN Centrum West
room

Private
18722 7605782 1476 39503013 NaN Amsterdam NaN Centrum West
room

17569 rows × 21 columns

In [ ]:

In [18]:
#13 Price column has large variation, so normalize the price column.

bnbscaled = bnb.copy()

# apply normalization techniques

for price in bnbscaled.price:

bnbscaled['price'] = (bnbscaled['price'] - bnbscaled['price'].min()) / (bnbscale


bnbscaled

Out[18]: room_id survey_id host_id room_type country city borough neighborhood

Shared De Pijp /
0 10176931 1476 49180562 NaN Amsterdam NaN
room Rivierenbuurt

Shared
1 8935871 1476 46718394 NaN Amsterdam NaN Centrum West
room

Shared
2 14011697 1476 10346595 NaN Amsterdam NaN Watergraafsmeer
room

Shared
3 6137978 1476 8685430 NaN Amsterdam NaN Centrum West
room

Shared De Baarsjes /
4 18630616 1476 70191803 NaN Amsterdam NaN
room Oud West

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

Private
18718 17789893 1476 47501089 NaN Amsterdam NaN Bijlmer Centrum
room

Private
18719 16877166 1476 67093870 NaN Amsterdam NaN Bijlmer Centrum
room

localhost:8888/nbconvert/html/Airbnb Case _SHIRIN MAHAJAN.ipynb?download=false 7/9


9/9/21, 10:35 AM Airbnb Case _SHIRIN MAHAJAN

room_id survey_id host_id room_type country city borough neighborhood

Private Geuzenveld /
18720 19859427 1476 29724632 NaN Amsterdam NaN
room Slotermeer

Private
18721 17132164 1476 115156569 NaN Amsterdam NaN Centrum West
room

Private
18722 7605782 1476 39503013 NaN Amsterdam NaN Centrum West
room

17569 rows × 21 columns

In [19]:
bnb['price'].mean()

Out[19]: 168.9025556377711

In [20]:
bnb['price'].std()

Out[20]: 110.98841382800018

In [21]:
#14 Change the price variable to three levels- Basic, medium and luxury

bnbscaled['price'].max()

Out[21]: 1.0

In [22]:
bnbscaled['price'].min()

Out[22]: 0.0

In [ ]:
conditions = [

(bnbscaled['price'] == 0),

(bnbscaled['price'] >0) & (airbscaled['price'] <=0.05 ),

(bnbscaled['price'] == 1)

values = ['Basic','medium','Luxury']

# create a new column and use np.select to assign values to it using our lists as ar
bnbscaled['Newprice'] = np.select(conditions,values)

# display updated DataFrame

bnbscaled.head()

In [23]:
#15 Drop the last 3 columns and save the data frame.

bnbscaled= bnbscaled.iloc[: , :-3]

In [24]:
bnbscaled

localhost:8888/nbconvert/html/Airbnb Case _SHIRIN MAHAJAN.ipynb?download=false 8/9


9/9/21, 10:35 AM Airbnb Case _SHIRIN MAHAJAN

Out[24]: room_id survey_id host_id room_type country city borough neighborhood

Shared De Pijp /
0 10176931 1476 49180562 NaN Amsterdam NaN
room Rivierenbuurt

Shared
1 8935871 1476 46718394 NaN Amsterdam NaN Centrum West
room

Shared
2 14011697 1476 10346595 NaN Amsterdam NaN Watergraafsmeer
room

Shared
3 6137978 1476 8685430 NaN Amsterdam NaN Centrum West
room

Shared De Baarsjes /
4 18630616 1476 70191803 NaN Amsterdam NaN
room Oud West

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

Private
18718 17789893 1476 47501089 NaN Amsterdam NaN Bijlmer Centrum
room

Private
18719 16877166 1476 67093870 NaN Amsterdam NaN Bijlmer Centrum
room

Private Geuzenveld /
18720 19859427 1476 29724632 NaN Amsterdam NaN
room Slotermeer

Private
18721 17132164 1476 115156569 NaN Amsterdam NaN Centrum West
room

Private
18722 7605782 1476 39503013 NaN Amsterdam NaN Centrum West
room

17569 rows × 18 columns

In [ ]:

localhost:8888/nbconvert/html/Airbnb Case _SHIRIN MAHAJAN.ipynb?download=false 9/9

You might also like