Scribd
Upload
9 views18 pages

Netflix Data Analysis 1691522070

The document presents an exploratory data analysis of a Netflix user dataset containing 2500 rows and 10 columns. It includes data cleaning steps, feature engineering, and visualizations to analyze subscription types, gender distribution, and age demographics. The analysis aims to provide insights into user behavior and subscription patterns across different countries and demographics.

Uploaded by

GOURAV KUMAR
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
9 views18 pages

Netflix Data Analysis 1691522070

The document presents an exploratory data analysis of a Netflix user dataset containing 2500 rows and 10 columns. It includes data cleaning steps, feature engineering, and visualizations to analyze subscription types, gender distribution, and age demographics. The analysis aims to provide insights into user behavior and subscription patterns across different countries and demographics.

Uploaded by

GOURAV KUMAR
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/ 18

netflix-data-analysis

July 30, 2023

1 Exploratory Data Analysis Netflix


[189]: import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
import seaborn as sns
import warnings
warnings.filterwarnings("ignore")

data = pd.read_csv("Netflix Userbase.csv")


np.random.seed(1)
print(f"In this dataset {data.shape[0]} rows and {data.shape[1]} columns")
data.sample(n=5)

In this dataset 2500 rows and 10 columns

[189]: User ID Subscription Type Monthly Revenue Join Date Last Payment Date \
1406 1407 Basic 14 31-10-22 04-07-23
297 298 Standard 14 03-11-22 27-06-23
1276 1277 Premium 11 30-09-22 04-07-23
2368 2369 Standard 11 24-09-22 13-07-23
438 439 Standard 12 23-10-22 27-06-23

Country Age Gender Device Plan Duration


1406 Canada 45 Female Smart TV 1 Month
297 Mexico 41 Male Tablet 1 Month
1276 Brazil 30 Male Smartphone 1 Month
2368 Australia 40 Female Smartphone 1 Month
438 Spain 37 Female Smart TV 1 Month

[190]: '''
we dont have userID so we can
that userID
'''
data.drop("User ID",axis=1,inplace=True)
data.head()

1
[190]: Subscription Type Monthly Revenue Join Date Last Payment Date \
0 Basic 10 15-01-22 10-06-23
1 Premium 15 05-09-21 22-06-23
2 Standard 12 28-02-23 27-06-23
3 Standard 12 10-07-22 26-06-23
4 Basic 10 01-05-23 28-06-23

Country Age Gender Device Plan Duration


0 United States 28 Male Smartphone 1 Month
1 Canada 35 Female Tablet 1 Month
2 United Kingdom 42 Male Smart TV 1 Month
3 Australia 51 Female Laptop 1 Month
4 Germany 33 Male Smartphone 1 Month

[191]: data.dtypes

[191]: Subscription Type object


Monthly Revenue int64
Join Date object
Last Payment Date object
Country object
Age int64
Gender object
Device object
Plan Duration object
dtype: object

[192]: data.isnull().sum()

[192]: Subscription Type 0


Monthly Revenue 0
Join Date 0
Last Payment Date 0
Country 0
Age 0
Gender 0
Device 0
Plan Duration 0
dtype: int64

[193]: #check the duplicate values

duplicates = data[data.duplicated()]

if duplicates.empty:
print("No duplicates Found!")
else:

2
print("Duplicates Found")
print(duplicates)

No duplicates Found!

[194]: '''
changing dates in the datetime format
'''
data["Join Date"] = pd.to_datetime(data["Join Date"])
data["Last Payment Date"] = pd.to_datetime(data["Last Payment Date"])

'''
Handling the datetime features
'''
data["Join Year "] = data["Join Date"].dt.year
data["Join Month "] = data["Join Date"].dt.month
data["Last Payment Year"] = data["Last Payment Date"].dt.year
data["Last Payment Month"] = data["Last Payment Date"].dt.month
data["Account till"] = (pd.to_datetime("today")-data["Join Date"]).dt.days
data.head()

[194]: Subscription Type Monthly Revenue Join Date Last Payment Date \
0 Basic 10 2022-01-15 2023-10-06
1 Premium 15 2021-05-09 2023-06-22
2 Standard 12 2023-02-28 2023-06-27
3 Standard 12 2022-10-07 2023-06-26
4 Basic 10 2023-01-05 2023-06-28

Country Age Gender Device Plan Duration Join Year \


0 United States 28 Male Smartphone 1 Month 2022
1 Canada 35 Female Tablet 1 Month 2021
2 United Kingdom 42 Male Smart TV 1 Month 2023
3 Australia 51 Female Laptop 1 Month 2022
4 Germany 33 Male Smartphone 1 Month 2023

Join Month Last Payment Year Last Payment Month Account till
0 1 2023 10 562
1 5 2023 6 813
2 2 2023 6 153
3 10 2023 6 297
4 1 2023 6 207

[195]: data.rename(columns={"Plan Duration":"Plan Duration(months)"},


inplace = True)

for i in range(len(data)):
data["Plan Duration(months)"] = data["Plan Duration(months)"][i][0]

3
data.head()

[195]: Subscription Type Monthly Revenue Join Date Last Payment Date \
0 Basic 10 2022-01-15 2023-10-06
1 Premium 15 2021-05-09 2023-06-22
2 Standard 12 2023-02-28 2023-06-27
3 Standard 12 2022-10-07 2023-06-26
4 Basic 10 2023-01-05 2023-06-28

Country Age Gender Device Plan Duration(months) Join Year \


0 United States 28 Male Smartphone 1 2022
1 Canada 35 Female Tablet 1 2021
2 United Kingdom 42 Male Smart TV 1 2023
3 Australia 51 Female Laptop 1 2022
4 Germany 33 Male Smartphone 1 2023

Join Month Last Payment Year Last Payment Month Account till
0 1 2023 10 562
1 5 2023 6 813
2 2 2023 6 153
3 10 2023 6 297
4 1 2023 6 207

[196]: """Adding new columns are filled with dummy values


based on previous features
"""
#Feature encoding
subscription_type_en = pd.get_dummies(data["Subscription␣
↪Type"],prefix="subscription type")

data = pd.concat([data, subscription_type_en],axis=1)

Country_en = pd.get_dummies(data["Country"],prefix="Country_type")
data = pd.concat([data, Country_en],axis=1)

Gender_en = pd.get_dummies(data["Gender"],prefix="Gender_type")
data = pd.concat([data, Gender_en],axis=1)

Device_en = pd.get_dummies(data["Device"],prefix="Device_type")
data = pd.concat([data, Device_en],axis=1)

#feature transform
data["Age Bins"] = pd.cut(data["Age"],
bins=[0,18,28,38,48,58,100],
labels=["<18","18-27","28-37","38-47","48-57","58+"])
data.head()

4
[196]: Subscription Type Monthly Revenue Join Date Last Payment Date \
0 Basic 10 2022-01-15 2023-10-06
1 Premium 15 2021-05-09 2023-06-22
2 Standard 12 2023-02-28 2023-06-27
3 Standard 12 2022-10-07 2023-06-26
4 Basic 10 2023-01-05 2023-06-28

Country Age Gender Device Plan Duration(months) Join Year \


0 United States 28 Male Smartphone 1 2022
1 Canada 35 Female Tablet 1 2021
2 United Kingdom 42 Male Smart TV 1 2023
3 Australia 51 Female Laptop 1 2022
4 Germany 33 Male Smartphone 1 2023

… Country_type_Spain Country_type_United Kingdom \


0 … 0 0
1 … 0 0
2 … 0 1
3 … 0 0
4 … 0 0

Country_type_United States Gender_type_Female Gender_type_Male \


0 1 0 1
1 0 1 0
2 0 0 1
3 0 1 0
4 0 0 1

Device_type_Laptop Device_type_Smart TV Device_type_Smartphone \


0 0 0 1
1 0 0 0
2 0 1 0
3 1 0 0
4 0 0 1

Device_type_Tablet Age Bins


0 0 18-27
1 1 28-37
2 0 38-47
3 0 48-57
4 0 28-37

[5 rows x 34 columns]

[197]: data.info()

<class 'pandas.core.frame.DataFrame'>

5
RangeIndex: 2500 entries, 0 to 2499
Data columns (total 34 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 Subscription Type 2500 non-null object
1 Monthly Revenue 2500 non-null int64
2 Join Date 2500 non-null datetime64[ns]
3 Last Payment Date 2500 non-null datetime64[ns]
4 Country 2500 non-null object
5 Age 2500 non-null int64
6 Gender 2500 non-null object
7 Device 2500 non-null object
8 Plan Duration(months) 2500 non-null object
9 Join Year 2500 non-null int64
10 Join Month 2500 non-null int64
11 Last Payment Year 2500 non-null int64
12 Last Payment Month 2500 non-null int64
13 Account till 2500 non-null int64
14 subscription type_Basic 2500 non-null uint8
15 subscription type_Premium 2500 non-null uint8
16 subscription type_Standard 2500 non-null uint8
17 Country_type_Australia 2500 non-null uint8
18 Country_type_Brazil 2500 non-null uint8
19 Country_type_Canada 2500 non-null uint8
20 Country_type_France 2500 non-null uint8
21 Country_type_Germany 2500 non-null uint8
22 Country_type_Italy 2500 non-null uint8
23 Country_type_Mexico 2500 non-null uint8
24 Country_type_Spain 2500 non-null uint8
25 Country_type_United Kingdom 2500 non-null uint8
26 Country_type_United States 2500 non-null uint8
27 Gender_type_Female 2500 non-null uint8
28 Gender_type_Male 2500 non-null uint8
29 Device_type_Laptop 2500 non-null uint8
30 Device_type_Smart TV 2500 non-null uint8
31 Device_type_Smartphone 2500 non-null uint8
32 Device_type_Tablet 2500 non-null uint8
33 Age Bins 2500 non-null category
dtypes: category(1), datetime64[ns](2), int64(7), object(5), uint8(19)
memory usage: 273.6+ KB

[198]: #check if which subscription is higher and subscription type


plt.figure(figsize=(22,20))
plt.subplot(1,2,1)
plt.title("Subscriptions Count")
sns.countplot(x="Subscription Type",
data=data,

6
color="green",hatch="/")
#apply text on plots
plt.text(0,1010,"Basic",fontsize=10
,fontweight="bold",
color="red")

plt.text(1,810,"Premium",fontsize=10
,fontweight="bold",
color="red")

plt.text(2,810,"Standard",fontsize=10,
fontweight="bold",
color="red")

plt.subplot(1,2,2)
sns.countplot(data=data,x="Country",
hue="Subscription Type",
palette="muted")

plt.title("Subscription Type by Country")


plt.xlabel("Country")
plt.ylabel("count")
plt.legend(title="subscription type")
plt.show()

7
[199]: #gender distribution
plt.figure(figsize=(8,6))
gender = data.Gender.value_counts()
index = ["Male","Female"]
colors = ["m","c"]
exp = [0.01,0.1]
values= gender.values.tolist()

plt.pie(values,
labels=index,
autopct="%.2f%%",
colors=colors,
explode=exp,
shadow=True,
startangle=80)
plt.title('Gender Distribution')

8
plt.show()

[200]: #age distribution


plt.figure(figsize=(12,10))
sns.histplot(data=data,
x="Age",
stat="count",
color="b",
edgecolor="cyan",
lw=2)
plt.title("Age Distribution")
plt.show()

9
[201]: plt.figure(figsize=(8,6))
sns.countplot(data=data,
x="Subscription Type",
hue="Age Bins",
edgecolor="blue",
lw=1
)
plt.title("Age by Subscription type")
plt.legend(loc="upper right")
plt.show()

10
[202]: #monthly revenue by country and device revenue monthly
plt.figure(figsize=(25,23))
plt.subplot(1,2,1)
#device by revenue
sns.barplot(x="Device",
y ="Monthly Revenue",
data=data,
edgecolor="blue")
plt.title("Monthly revenue each device")
#country by revenue
plt.subplot(1,2,2)
sns.barplot(x="Country",
y="Monthly Revenue",
data=data,
edgecolor="blue")
plt.title("Monthly revenue each country")
plt.show()

11
[203]: #People of this age are using this device
sns.countplot(x="Age Bins",
hue="Device",
data=data,
edgecolor="blue",
lw=1)
plt.title("Age by device")
plt.ylabel("Device")
plt.show()

12
[204]: #monthly revenue by device
sns.barplot(x="Device",
y="Monthly Revenue",
data=data,
edgecolor="blue"
)
plt.show()

13
[205]: #percentage of subscription
x = data["Subscription Type"].value_counts(ascending=False)
index = [i for i in x.index]
var = [0,0.02,0.1]
plt.figure(figsize=(8,6))
plt.pie(x.values,
labels=index,
autopct="%.2f%%",
explode=var,
shadow=True,
startangle=90)
plt.show()

14
[206]: #Subscription revenue every month
sns.barplot(x="Subscription Type",
y="Monthly Revenue",
hue="Device",
data=data,
edgecolor="blue",
lw=1)
plt.legend(loc="upper right")
plt.title("Subscription Revenue Monthly")
plt.show()

15
[207]: #subscription type by plan duration
plt.figure(figsize=(14,10))
plt.subplot(1,2,1)
sns.countplot(data=data,
x="Subscription Type",
hue="Plan Duration(months)",
palette="muted")
plt.title("Subscription Type by plan Duration")
plt.ylabel("Plan Duration")
#plan duration of ages
plt.subplot(1,2,2)
sns.countplot(data=data,
x="Age Bins",
hue="Plan Duration(months)",
palette="muted")
plt.legend(loc="upper right")
plt.title("Plan Duration Age Bins")
plt.ylabel("plan duration")
plt.show()

16
[208]: #Monthly revenue
plt.figure(figsize=(10,6))
sns.lineplot(x="Join Month ",
y="Monthly Revenue",
data=data,
estimator= "mean",
color="blue")
plt.title("Monthly Revenue Over Time")
plt.show()

17
2 End!

18

You might also like