GeekFinder is a smart ranking system from where you can find peers who share similar learning paths as you, to collaborate and work together! This is intended for College Students including both Undergrads and Postgrads. We utilized the concept of a Content-Based Recommender to accomplish this.
Try GeekFinder here
Demo:
- Python
- Flask
- Scikit-learn
- Pandas
- Numpy
- Seaborn
- HTML
- CSS
In the end, the web app was deployed to Heroku.
- Recommender Systems 101
- Applying ML in Dating Apps
- Deeper look into Content-Based Recommendation
- LightFM - A Hybrid Recommendation Algorithm in Python
This project was ideated, developed and deployed end-to-end within just 6 weeks with a team of four members from the Batch 1 of Crework's Build With Us Program.