Project Recommendation System

The goal of this project is to make a simple and light recommendation system that helps in recommending similar circuits to the users in the most efficient way possible. From the database, the User ID, Project ID, Project description, the number of Views, and the number of stars will be extracted and saved in a CSV file which would then be loaded into the program. The description is used in clustering similar circuits together and then the views and stars (by choosing some weights) would be used to re-rank within a cluster. Feature extraction is done using the TF-IDF encoding and a text embedding is made on which a suitable clustering algorithm would be used in order to group similar circuits together and sent for re-ranking. A content boosted approach can also be used while re-ranking and using a similarity measure while displaying the circuits to users in real-time. The project later focuses on integration with the Circuit Verse website by helping with the UI of the recommendation system.