App Board

Dashboard and Stats of Applications available on the Apple App Store.

Comment

Inspiration

Being a developer myself, I know that the most challenging part of the development process is neither the development of the application nor the testing of it. Determining what and when to implement is the most troublesome. There are hundreds of applications available on the App Store, both Apple and Android, and not all the developers who made them are prosperous. Imagine if you knew beforehand, in what period, which app would be most thriving? What Genre of application is going to be trending the most? What price to sell your app at to maximize the profits?

What it does

AppBoard is an analytical dashboard that compiles the insights found in the 736K IOS Apps Dataset. Besides the visualizations available in the dashboard, the Notebooks folder includes all the analysis and visualizations done in the jupyter notebook, along with Data Cleaning and pre-processing.

Technical Implementation

The dashboard follows the UI aesthetic of the bridge web application, and due to the integration of Bootstrap, it is responsive on mobile and smaller devices. Dash and plotly were used for implementation of the dashboard which, was later deployed to the Heroku server.

The first step after obtaining the dataset was pre-processing, achieved using numpyand pandas packages of python. Removal of Null Values and unwanted attributes took place, and the Date-Time format was also corrected.

After pre-processing, analysis and visualization began. Libraries like Seaborn and Matplotlib were utilized for the visualization part.

Challenges faced

I had no experience in developing dashboards with Plotly and Dash. Also, I did not know about the deployment of a dashboard on the Heroku server.

Accomplishments

With minimal expertise in Dash, Plotly, and Heroku, I created a Live Dashboard after reading documentations, articles from "towards data science", and watching tutorial videos.

What I learned

  • Dash and Plotly (Dashboard Development)
  • Exploratory Data Analysis (Numpy, Pandas, Seaborn, Matplotlib)

What's next for App Board

  • Integrating Callbacks (Multiple Inputs and Outputs) for a better interactive dashboard.
  • Adding new Dash Core and HTML Components for a better understanding of the data.
  • Supplementary Data like application size, update size, etc., for better insights.
  • Machine Learning Model to predict the price of any app before deploying to the Apple App Store.
Share this project:

Updates