Skip to content

A graph-based personalized LeetCode question recommender that uses probabilistic reasoning and topic modeling for skill enhancement.

License

Notifications You must be signed in to change notification settings

amri-tah/LeetPath

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

LeetPath: A Graph-Based LeetCode Question Recommender πŸ§­πŸ”

image

LeetPath is a personalized question recommendation system designed for LeetCode users. Using graph-based structures, topic modeling, and Markov Random Field, the system analyzes user interaction, question similarity, and topic relevance to suggest the most appropriate questions for continued skill improvement. πŸ“ˆ

Demo πŸŽ₯

leetpathdemo1.mp4

Features 🌟

  • πŸ” Personalized Recommendations: Suggests questions based on user activity and skills.
  • 🧠 Topic Modeling: Groups questions by related topics for better understanding.
  • πŸ“Š Interactive Dashboard: Displays user stats and recommended questions.
  • ⚑ Real-Time Performance: Fast and efficient recommendations using graph-based algorithms and belief propagation.

Tech Stack πŸ› οΈ

Tech_Stack

  • Frontend: Next.js with Tailwind CSS for a responsive and interactive UI.
  • Backend: Flask for API endpoints for the model and Go for fetching user data.
  • Database: MongoDB for storing user data.
  • Authentication: Firebase for secure user login and management.
  • GraphQL: For efficient and flexible data querying.

Deployment and Hosting πŸ’»

  • The recommendation engine is deployed on Google Cloud Platform using App Engine.
  • The backend code is deployed on OnRender.
  • The frontend is hosted on Vercel.

How It Works πŸ“Š

  • Question Similarity: Content-based filtering using TF-IDF and cosine similarity to recommend questions based on their content similarity.
  • Topic Modeling: Grouping questions by latent topics using a custom topic modeling algorithm (similar to Latent Dirichlet Allocation).
  • Markov Random Field (MRF): Models relationships between questions, accounting for user engagement, difficulty, and question similarities.
  • Belief Propagation: Used to refine potential values in the MRF and improve recommendation accuracy.

Screenshots

landing Screenshot 2024-11-21 215125 Screenshot 2024-11-21 215155 recommender page 1 profile

Contributors

@VishalTheHuman @amri-tah @yeager209904 @GiriPrasath017
Vishal S Amritha Nandini Anerud Thiyagarajan Giri Prasath R

Contributing 🌟

We welcome contributions to enhance the functionality of LeetPath! If you have ideas or improvements, please submit a pull request . πŸš€

License πŸ“œ

This project is licensed under the MIT License. See the LICENSE file for more details . πŸ“„

Contact πŸ“§

For any queries or support, please contact us at amrithanandini2003@gmail.com or vishalatmadurai@gmail.com. We're here to help you!πŸ“¬

Thank you for using LeetPath! Let's elevate your LeetCode experience together. πŸš€πŸ’»

About

A graph-based personalized LeetCode question recommender that uses probabilistic reasoning and topic modeling for skill enhancement.

Topics

Resources

License

Stars

Watchers

Forks

Contributors 5