The Android App Market on Google Play

Mobile apps are everywhere. They are easy to create and can be lucrative. Because of these two factors, more and more apps are being developed. In this project, you will do a comprehensive analysis of the Android app market by comparing over ten thousand apps in Google Play across different categories. You'll look for insights in the data to devise strategies to drive growth and retention. The goal of this notebook is to illustrate how we managed to apply data cleaning processes on the two given datasets and to find correlations between its attributes and how they can affect each other. Moreover, we will discuss some statistics and figures related to the datasets.

Columns description:

App: Application name

Category: Category the app belongs to

Rating: Overall user rating of the app (as when scraped)

Reviews: Number of user reviews for the app (as when scraped)

Installs: Number of user downloads/installs for the app (as when scraped)

Type: Paid or Free

Content Rating: Age group the app is targeted at - Children / Mature 21+ / Adult

Genres: An app can belong to multiple genres (apart from its main category). For eg, a musical family game will belong to Music, Game, Family genres.

Last Updated: Date when the app was last updated on Play Store (as when scraped)

Current Ver: Current version of the app available on Play Store (as when scraped)

Android Ver: Min required Android version (as when scraped)

Our Data Processing Pipeline

A. Investigating different app categories

1- How many unique categories are in the dataset?

First, we removed the duplicates in the dataset. After counting the number of unique categories, we found that there are 34 different categories in the dataset. Then we counted the number of apps in each category and we found that there is a category called "1.9" that contains only one app. We searched for this category and we realized that it is related to the "LIFESTYLE" category. So our dataset now has 33 different categories.

Which of these categories have the fewest number of apps?

In the figure above, we can observe that the categories BEAUTY, COMICS, PARENTING, AUTO_AND_VEHICLES, LIBRARIES_AND_DEMO, WEATHER HOUSE_AND_HOME and EVENTS ART_AND_DESIGN have the fewest number of apps which does not exceed 100 apps.

3- Which of these categories have the highest number of apps?

In the figure above, we can also observe that the categories FAMILY, GAMES and TOOLS have the highest number of apps which exceeds 500 apps. #### 4- What type of categories do people mostly install?

In the figure above, we can observe that GAME, COMMUNICATION and TOOLS are the top three categories that people mostly install. GAME comes first with more than 13 Billion installs, COMMUNICATION comes second with more than 11 billion installs and TOOLS comes third with more than 8 Billion installs.

B. Exploring Free and Paid apps

1- Which type of apps are mostly downloaded ? (whether free or paid depending on num. of installs ?

The chart above represents the percentage of free installed apps compared to the paid ones. This comparison seems to be unexpected as the difference between the two values is huge. However, in the chart below we compared the number of free apps to the number of paid ones in the dataset and we found out that the difference is huge as well.

2- How many categories contain free and paid apps together?

29 out of 33 categories contain both free and paid apps, the other 4 categories contain only free apps.

3- How many free and paid applications?

So it makes sense that the difference between the number of installs of both types is huge.

4- What are the categories that contain the most expensive apps?

The figure above shows the categories that have the most expensive apps in the dataset. LIFESTYLE, FINANCE and FAMILY have the most expensive apps which nearly cost $400.

5- Which apps in the above categories have the more than 1000 installs?

We observed that the most expensive application in the three categories mentioned above is the same application but with different editions. For example, all the applications contains "I'm rich" or " i am rich" concatinated with different words like "Pro" or "Premium". This observation took our attention as the number of installs of each variant is more than 1000 installs given that each app costs nearly 400$.

6- How many categories contain free and paid apps together?

29 out of 33 categories contain both free and paid apps, the other 4 categories contain only free apps.

Investigating App categories

1. What are the top rated categories ?

We computed the average rating of all categories using the rating of each application in each category to determine the top rated categories. We found that the top five categories are Events, Art and Design, Books and Reference, Education and Personalization. However the top installed categories are Game, Communication, Tools, Productivity and Social, in which there is no matching between top rated categories and top installed categories. The aforementioned result was expected because it is logically that the category got high number of installed might not get a high rating as this means that people discover bugs and features they prefer not listed in these apps in the top installed categories.

2. Representing the rating values by Boxplot diagram; we can view the minimum, maximum, mean and outliers

For example, for the 'ART AND DESIGN'; the maximum was at 5.0, the minimum was at 3.25 and the mean was slightly below 4.5 and for the 'BOOKS AND REFERENCE'; the maximum was at 5.0, the minimum was at 3.7 , the mean was slightly above 4.0 and showing multiple outliers below the minimum. and similarly for the rest of the values.

Final Analysis

We learnt from our data that there are 33 category and the categories of the top applications mostly installed which are Games and Communication. The aforementioned result was expected as the users mostly search for the type of application that entertain them and help them to connect to other people. Another finding is that the percentage of installed free apps is 99.9%, however even this percentage is considered high it might be reasonable to us as we found the percentage of free app in our dataset is 99.2% . We used different statistical visualizations to answer our questions like bar and pie charts .