The Google Play Store apps data analysis provides enough potential to drive apps making businesses to succeed. Actionable stats can be drawn for developers to work on and capture the Android market. The data set that I have taken in this article is a web scrapped data of 10 thousand Playstore applications to analyze the android competition.
In this Article I will do some Exploratory Data Analysis on the Google Play Store apps data with Python.
Lets start with importing the libraries
import numpy as np import pandas as pd import matplotlib.pyplot as plt import seaborn as sns
Download the Data set
df = pd.read_csv('googleplaystore.csv')Lets see at some insights of the data
df.info()
<class 'pandas.core.frame.DataFrame'> RangeIndex: 10841 entries, 0 to 10840 Data columns (total 13 columns): # Column Non-Null Count Dtype --- ------ -------------- ----- 0 App 10841 non-null object 1 Category 10841 non-null object 2 Rating 9367 non-null float64 3 Reviews 10841 non-null object 4 Size 10841 non-null object 5 Installs 10841 non-null object 6 Type 10840 non-null object 7 Price 10841 non-null object 8 Content Rating 10840 non-null object 9 Genres 10841 non-null object 10 Last Updated 10841 non-null object 11 Current Ver 10833 non-null object 12 Android Ver 10838 non-null object dtypes: float64(1), object(12) memory usage: 1.1+ MB
Exploratory Data Analysis on Google Play Store
Let’s take a look on all the category
# Category cat = df.Category.unique() cat
array(['ART_AND_DESIGN', 'AUTO_AND_VEHICLES', 'BEAUTY',
'BOOKS_AND_REFERENCE', 'BUSINESS', 'COMICS', 'COMMUNICATION',
'DATING', 'EDUCATION', 'ENTERTAINMENT', 'EVENTS', 'FINANCE',
'FOOD_AND_DRINK', 'HEALTH_AND_FITNESS', 'HOUSE_AND_HOME',
'LIBRARIES_AND_DEMO', 'LIFESTYLE', 'GAME', 'FAMILY', 'MEDICAL',
'SOCIAL', 'SHOPPING', 'PHOTOGRAPHY', 'SPORTS', 'TRAVEL_AND_LOCAL',
'TOOLS', 'PERSONALIZATION', 'PRODUCTIVITY', 'PARENTING', 'WEATHER',
'VIDEO_PLAYERS', 'NEWS_AND_MAGAZINES', 'MAPS_AND_NAVIGATION',
'1.9'], dtype=object)So we got 34 category on this dat aset, let’s see which one is the famous category
plt.figure(figsize=(12,12)) most_cat = df.Category.value_counts() sns.barplot(x=most_cat, y=most_cat.index, data=df)
So, there is around 2000 app with family category, followed by game category with 1200 app. And this ‘1.9’ Category, i don’t know what it is, but it only had 1 app so far, so its not visible on the graph.
Let’s look at the rating, and what kind of correlation share between category and rating.
# Rating df.Rating.unique()
array([ 4.1, 3.9, 4.7, 4.5, 4.3, 4.4, 3.8, 4.2, 4.6, 3.2, 4. ,
nan, 4.8, 4.9, 3.6, 3.7, 3.3, 3.4, 3.5, 3.1, 5. , 2.6,
3. , 1.9, 2.5, 2.8, 2.7, 1. , 2.9, 2.3, 2.2, 1.7, 2. ,
1.8, 2.4, 1.6, 2.1, 1.4, 1.5, 1.2, 19. ])There we had a null values, I am going to leave it as it is. And a 19 for rating is not possible, so i assume it’s a ‘1.9’. So let’s change it and see the distribution value on rating column.
df['Rating'].replace(to_replace=[19.0], value=[1.9],inplace=True) sns.distplot(df.Rating)
Most of the rating is around 4. Let’s see how rating is distributed by category column.
g = sns.FacetGrid(df, col='Category', palette="Set1", col_wrap=5, height=4) g = (g.map(sns.distplot, "Rating", hist=False, rug=True, color="r"))
By the horizontal is the rating value, and vertically is quantity of the rating.
# Mean Rating plt.figure(figsize=(12,12)) mean_rat = df.groupby(['Category'])['Rating'].mean().sort_values(ascending=False) sns.barplot(x=mean_rat, y=mean_rat.index, data=df)
And this is the average of rating by category, family and game has a lot of quantity causing the low on average rating, on the other side event has the highest average rating by category.
Next is reviews, review sometime can measure the app popularity. The more reviews, the better.
# Reviews df.Reviews.unique()
Output-
array([‘159’, ‘967’, ‘87510’, …, ‘603’, ‘1195’, ‘398307’], dtype=object)
# inside review there is a value with 3.0M with M stand for million, lets change it so it can be measure as float
Reviews = []
for x in df.Reviews:
x = x.replace('M','00')
Reviews.append(x)
Reviews = list(map(float, Reviews))
df['reviews'] = Reviews
sns.distplot(Reviews)
This graph is the distribution of total reviews on each app.
g = sns.FacetGrid(df, col='Category', palette="Set1", col_wrap=5, height=4) g = (g.map(plt.hist, "Reviews", color="g"))
This graph is the correlation between category and reviews, Family and game category had a lot of reviews.
Some app also almost had no review at all, like event, beauty, medical, parenting and more. it is interesting Event app has a high rating but rare review on it.
# Total reviews plt.figure(figsize=(12,12)) sum_rew = df.groupby(['Category'])['reviews'].sum().sort_values(ascending=False) sns.barplot(x=sum_rew, y=sum_rew.index, data=df)
Showing the amount of total reviews.
# Mean reviews plt.figure(figsize=(12,12)) mean_rew = df.groupby(['Category'])['reviews'].mean().sort_values(ascending=False) sns.barplot(x=mean_rew, y=mean_rew.index, data=df)
This is the average of reviews on each category. Let’s move on to next column, installs.
# Installs df.Installs.unique()
array(['10,000+', '500,000+', '5,000,000+', '50,000,000+', '100,000+',
'50,000+', '1,000,000+', '10,000,000+', '5,000+', '100,000,000+',
'1,000,000,000+', '1,000+', '500,000,000+', '50+', '100+', '500+',
'10+', '1+', '5+', '0+', '0', 'Free'], dtype=object)Now i’m going to transform this column into float as well like review. First we need to change the 0 and Free value to 0+.
Next we need to replace the ‘,’ value and discard the + sign form the value.
df['Installs'].replace(to_replace=['0', 'Free'], value=['0+','0+'],inplace=True)
Installs = []
for x in df.Installs:
x = x.replace(',', '')
Installs.append(x[:-1])
Installs = list(map(float, Installs))
df['installs'] = Installs
sns.distplot(Installs)
Distributed value of Install on each category.
g = sns.FacetGrid(df, col='Category', palette="Set1", col_wrap=5, height=4) g = (g.map(plt.hist, "installs", bins=5, color='c'))
# Total Installs plt.figure(figsize=(12,12)) sum_inst = df.groupby(['Category'])['installs'].sum().sort_values(ascending=False) sns.barplot(x=sum_inst, y=sum_inst.index, data=df)
# Mean Install plt.figure(figsize=(12,12)) mean_ints = df.groupby(['Category'])['installs'].mean().sort_values(ascending=False) sns.barplot(x=mean_ints, y=mean_ints.index, data=df)
The Type column, let’s check if the app is free or paid.
# Type for category df.Type.unique()
array(['Free', 'Paid', nan, '0'], dtype=object)
There is 0 and null value, let’s change them to free.
df['Type'].replace(to_replace=['0'], value=['Free'],inplace=True)
df['Type'].fillna('Free', inplace=True)print(df.groupby('Category')['Type'].value_counts())
Type_cat = df.groupby('Category')['Type'].value_counts().unstack().plot.barh(figsize=(10,20), width=0.7)
plt.show()Category Type
1.9 Free 1
ART_AND_DESIGN Free 62
Paid 3
AUTO_AND_VEHICLES Free 82
Paid 3
...
TRAVEL_AND_LOCAL Paid 12
VIDEO_PLAYERS Free 171
Paid 4
WEATHER Free 74
Paid 8
Name: Type, Length: 64, dtype: int64
So again, family category has the most free and paid app on the google play store. We can see social app is always free, like entertainment, event, education, comic, and more.
The medical has a high amount of paid app considering quantity of medical app is not much.
Last is the version of android you should have before accessing the app.
# Android Version df['Android Ver'].unique()
array(['4.0.3 and up', '4.2 and up', '4.4 and up', '2.3 and up',
'3.0 and up', '4.1 and up', '4.0 and up', '2.3.3 and up',
'Varies with device', '2.2 and up', '5.0 and up', '6.0 and up',
'1.6 and up', '1.5 and up', '2.1 and up', '7.0 and up',
'5.1 and up', '4.3 and up', '4.0.3 - 7.1.1', '2.0 and up',
'3.2 and up', '4.4W and up', '7.1 and up', '7.0 - 7.1.1',
'8.0 and up', '5.0 - 8.0', '3.1 and up', '2.0.1 and up',
'4.1 - 7.1.1', nan, '5.0 - 6.0', '1.0 and up', '2.2 - 7.1.1',
'5.0 - 7.1.1'], dtype=object)Now I am going to group it to 1 till 8 version of android. Change the null value to 1.0.
df['Android Ver'].replace(to_replace=['4.4W and up','Varies with device'], value=['4.4','1.0'],inplace=True)
df['Android Ver'].replace({k: '1.0' for k in ['1.0','1.0 and up','1.5 and up','1.6 and up']},inplace=True)
df['Android Ver'].replace({k: '2.0' for k in ['2.0 and up','2.0.1 and up','2.1 and up','2.2 and up','2.2 - 7.1.1','2.3 and up','2.3.3 and up']},inplace=True)
df['Android Ver'].replace({k: '3.0' for k in ['3.0 and up','3.1 and up','3.2 and up']},inplace=True)
df['Android Ver'].replace({k: '4.0' for k in ['4.0 and up','4.0.3 and up','4.0.3 - 7.1.1','4.1 and up','4.1 - 7.1.1','4.2 and up','4.3 and up','4.4','4.4 and up']},inplace=True)
df['Android Ver'].replace({k: '5.0' for k in ['5.0 - 6.0','5.0 - 7.1.1','5.0 - 8.0','5.0 and up','5.1 and up']},inplace=True)
df['Android Ver'].replace({k: '6.0' for k in ['6.0 and up']},inplace=True)
df['Android Ver'].replace({k: '7.0' for k in ['7.0 - 7.1.1','7.0 and up','7.1 and up']},inplace=True)
df['Android Ver'].replace({k: '8.0' for k in ['8.0 and up']},inplace=True)
df['Android Ver'].fillna('1.0', inplace=True)print(df.groupby('Category')['Android Ver'].value_counts())
Type_cat = df.groupby('Category')['Android Ver'].value_counts().unstack().plot.barh(figsize=(10,18), width=1)
plt.show()Category Android Ver
1.9 1.0 1
ART_AND_DESIGN 4.0 51
2.0 9
1.0 2
3.0 2
..
WEATHER 4.0 38
1.0 26
2.0 10
5.0 7
3.0 1
Name: Android Ver, Length: 200, dtype: int64
Also, read – 10 Machine Learning Projects to Boost your Portfolio
This dataset contains a good set of possibilities, to work more on the business values and leaving with a positive impact. This work is not restricted to until the exploration of this article. You can explore more interesting facts and figures using this article.
