Airtel Whack
May 21, 2017
Team Ignited_Minds presented a Recommendation and Prediction Engine named Suvidha at Airtel Whack Hackathon.
Suvidha is a data analysis tool that:
Studies user’s usage patterns.
Recommends optimal plans based on the pattern.
Provides Airtel an understanding about:
How a package is performing ?
How it can be improved ?
Methodology :
Study usage for past 6 months.
Classify usage stats into plan preferences.
Add user’s choice to existing database of usage patterns.
Admin views detailed analysis of user data on a gross level.
Design market strategies based on the usage patterns.
Primary use cases of Airtel.in / MyAirtel :
Checking remaining data, balance
Recharging
Improve the use case by incorporating Suvidha into existing services:
Provide a customised suggestion that impacts transparency
Saves money directly
Tech Stack
Python (NumPy For GaussianNB classification)
MongoDB
Flask (RESTful API)
REST API can be extended to app or any other platform
Check out the Airtel Slides for more info on how we made the recommendation engine. and the product demo here Airtel Suvidha Demo.
Keep tuned for more stories from my hackathon series.
Credits - Aditi Prerna (Writer & Editor). She is writing and compiling my stories at rjrajivjha.