Course Recommendation Web App An intelligent course recommendation system that personalizes learning paths based on user interests, course completion progress, and interaction history. The app dynamically recommends relevant courses, skill-based roadmaps, and related advanced technologies to enhance user upskilling journeys.
Key Features:
- Personalized recommendations via hybrid recommendation engine (content-based + user progress)
- Adaptive roadmap that adjusts suggestions based on course completion percentages
- Backend API with Flask and PostgreSQL for user and enrollment management
- React.js frontend for course selection, progress tracking, and dynamic recommendation feed
- Dockerized multi-container architecture for easy deployment and scaling
Unlike standard recommendations, our system adapts based on each user's course completion progress. This allows for more relevant and timely suggestions, encouraging continued learning.
-
Why it’s different: Goes beyond course history — it uses how much of a course the user has completed.
-
Why included: To guide users through a smart, personalized learning path.
-
Impact: Increases engagement by recommending courses aligned with current skill level and interest.
-
Responsible Component:
Backend API: /recommend route
Logic: generate_recommendations() in Recommend.py
Progress Tracking: /update_progress endpoint + enrollments table
- Recommendation Feed : 12 courses, 5 of More Completion Progress, 4 of Less Completeion Progress, 3 other subject course (for exploratory part).
- Docker and Docker Compose installed on your machine
-
Clone the repository:
git clone https://github.com/IcodeG00D/Course-Recommendation-Web-App.git cd Course-Recommendation-Web-Appbuild containers in root folder directory where docker-compose.yaml should be there.
-
Ensure your
.envor environment variables are properly set for database credentials (optional if overridden in Docker Compose). -
Build and start Docker containers:
docker-compose up --build
Incase:(issue occurs):
- Start the Docker Desktop App first, if Container shows any error while building.
- Try removing old volumes , and try again.
docker-compose down -v
- Look for Docker-compose.yaml file should be there
-
Access the application:
- Frontend:
http://localhost:3000(For UI CLICK HERE) - Backend API:
http://localhost:5000
- Frontend:
For detailed step instruction please refer to attached Installation manual
USER--> SUBJECT-->COURSE-->ENROLL-->UPDATE PROGRESS-->RECOMMEND FEED-->ENROLL NEW COURSE-->MY COURSES-->UPDATE PROGRESS/DELETE COURSE--->DATA IN DB
-
User Registration / Identification → User is identified (via /store_user) by name when they enter the platform.
-
Browse Subjects → Available subjects are fetched (/subjects) to let the user explore interest areas.
-
Select & View Courses by Subject → User selects a subject to view top courses (/top-courses/).
-
Enroll in a Course → User enrolls in a course (/enroll_course) — data is stored in enrollments.
-
Update Course Progress → As the user makes progress, updates are sent via /update_progress.
-
Receive Personalized Recommendations → Recommendations adapt in real-time via /recommend based on the progress and interest (generate_recommendations() logic).
-
Enroll in New Recommended Course → User selects another course from the feed and enrolls again.
-
My Courses View → User can view all enrolled courses and current progress (/merged-enrollments/ or /enrollments/).
-
Update Progress or Unenroll → User can: Continue progress and send updates (/update_progress) Or unenroll from a course (/unenroll_course)
-
Data Stored in PostgreSQL → All user activity is tracked in users and enrollments tables for personalized learning paths and future recommendations. Use the frontend React app to register, select courses, and track progress.The backend API handles recommendations based on course completion percentages and user input.
-
Example backend endpoints include:
POST /recommend— Get personalized course recommendationsPOST /enroll_course— Enroll a user in a coursePOST /update_progress— Update course progress percentageGET /enrollments/<user_name>— Fetch user enrollments and progress Refer section 8.5 in attatched Offical Documents
environment:
DB_HOST: db
DB_PORT: 5432
DB_NAME: coursedb
DB_USER: myuser
DB_PASS: mypass- The PostgreSQL container uses the above credentials to initialize the database.
- Backend connects to
dbhostname (Docker network alias).
- Backend: Python, Flask, psycopg2, scikit-learn (TF-IDF & cosine similarity)
- Frontend: React.js
- Database: PostgreSQL (via official Docker image)
- Containerization: Docker, Docker Compose
- Data Processing: pandas, numpy.
- Dataset: Udemy-Courses
For more info refer attatched Documentation.
This project uses a three-container setup orchestrated via docker-compose.yml:
- db: PostgreSQL database container
- backend: Flask API container
- frontend: React.js UI container
Run:
docker-compose up --buildThis will build images (if not already built) and start all containers with networking configured.
docker-compose down
here you can select the courses which are sorted on the basis of popularity (most enrolled).
In Img F you can see all the enrolled courses are been visible in the Database Table which means its updating the Table in Real time.
And here comes the homefeed with recommending all the courses on the basis of,
- Enrolled Course (Title and Subject).
- Most completed course (User showing interest).
- 12 courses, 5 of More Completion Progress, 4 of Less Completeion Progress, 3 other subject course (for exploratory part).
- Fork the repo and create feature branches for enhancements or fixes
- Follow PEP8 for Python backend and standard React.js practices on frontend
- Write descriptive commit messages
- Open pull requests with clear descriptions and testing instructions
- Report issues or feature requests on GitHub Issues
- Developer: Kushagra Pathak (GitHub: IcodeG00D)
- Dataset by AndrewMVd on Kaggle
- Inspired by hybrid recommendation system concepts combining content-based filtering and user progress metrics
Q: Backend fails to connect to the database?
A: Make sure the Docker db service is running and environment variables match in docker-compose.yml.
Q: How to update course progress?
A: Use the /update_progress API endpoint via frontend or direct API call with user_name, course, and progress fields.
Q: Recommendations seem irrelevant? A: Ensure course titles in user progress data exactly match dataset course titles. Dataset loading must be successful.
Q: How to add more courses?
A: Update the udemy_courses.csv dataset and rebuild the backend container.
Thank you , Hope you liked the Project .