Overview

Participated in Open IIT Data Analytics Hackathon 2021, sponsored by Brillio, where our team developed a predictive model to estimate the popularity of 4000+ songs to assist a media company in optimizing its bidding strategy for maximizing revenue.
We achieved 1st position out of 48 teams.

Project Details

• Built a Voting Classifier combining Random Forest, Support Vector Classifier, and Gradient Boosting.
• Achieved 91.2 % accuracy on the test set.
• Predicted optimal song bidding strategy, generating a simulated revenue of 10,500 units, which was 85 % of the theoretical maximum possible.
• The approach balanced interpretability and ensemble performance, making it suitable for real-time music analytics.

Highlights

🏆 Position: 1st / 48 teams
📍 Event: Open IIT Data Analytics, May 2021
🎯 Organizer: Brillio & IIT Kharagpur Inter Hall Tech Events

Tech Stack

Python · scikit-learn · Pandas · Matplotlib · NumPy

Links

• Event Page (add link if available)
• GitHub Repository (optional)