Recommendation System
Project Code: RSXX
Embedded Files
Mohd Taiyab Khan | Sabah Anwar Azmi
Revolutionizing Education with Personalized Learning: Our Recommendation System for Padhaai App's BITS Feature
Revolutionizing Education with Personalized Learning: Our Recommendation System for Padhaai App's BITS Feature
Background
Gurucool's Padhaai app is a platform for students to access educational content in an engaging and interactive way. To enhance the user experience, a recommendation system was developed for the BITS (Briefly, Interesting and Teachable Segments) feature, which is similar to Instagram's Reels.
The initial approach to the recommendation system was popularity-based, where only popular BITS were displayed to the users. The next step was user-based, where the system recommended BITS based on the user's past behavior. This approach provided a more personalized recommendation to the users.
The latest approach is the hybrid recommendation system, which combines both popularity-based and user-based approaches to provide even more personalized recommendations. The system was developed using Python and various libraries, and the pickel file was generated and connected to the database to make it real-time.
The hybrid recommendation system provides a personalized learning experience, making learning fun and engaging. With its efficient algorithms, the system ensures that users are provided with the most relevant and interesting BITS, enhancing their overall learning experience.
Resources
Python
Jupyter Notebook
User Data from APIs & Database
Machine Learning Algorithms
Team Updates
Research on Recommendation System.
Get and Preprocessed basic data.
Building a Popularity based Recommendation System.
Evaluation of Popularity.
Research on User Based.
Get the Database access and extracted all the necessary data.
Preprocessed the data and build a system to rate the BITS.
Building a User based Recommendation System and making it real time with Database.
Model Evaluation
Embedding it to our platform.
Plan of Action
Define requirements and goals of the recommendation system.
Gather and preprocess data for training the model.
Train and test various recommendation algorithms.
Implement the chosen algorithm in a production environment.
Monitor and continuously improve the system's performance.
Tech Inclusions
Getting the Data
Implementing the final version of Recommendation System built by RD-2 Team.
Future Plan
Planning to Built Hybrid Recommendation System.
Page updated
Google Sites
Report abuse