Multimodal E-learning Recommender System

Overview

The Multimodal E-learning Recommender System is an innovative project designed to deliver dynamic and personalized course recommendations by leveraging Neo4j graph databases and advanced machine learning algorithms. This project represents relationships between entities such as courses, students, instructors, and skills to provide insightful and tailored recommendations.

Features

• Graph-based Relationships: Using Neo4j to model relationships between courses, instructors, categories, and skills.
• Machine Learning Models: Includes hybrid recommendation models to generate personalized course suggestions.
• Community Detection: Implements the Louvain algorithm for detecting communities in the dataset.
• Python-Based Analysis: Detailed course distribution insights using Python libraries.

Project Objectives

1. Build a robust recommendation system for e-learning platforms.
2. Utilize multimodal graphs for representing course and user interactions.
3. Perform advanced data analysis to improve recommendation accuracy.

Technologies Used

• Programming Languages: Python
• Graph Database: Neo4j
• Machine Learning Algorithms: Hybrid recommendation models
• Community Detection Algorithm: Louvain
• Visualization Tools: Python libraries like Matplotlib and Seaborn

Dataset Details

The project dataset includes the following columns:

• Course Details: Name, Short Intro, Category, Subcategory, Course Type
• Attributes: Language, Skills, Rating, Number of Students, Duration
• Other: Hosting Website, Instructor

Installation

1. Clone the repository:

   git clone https://github.com/aarush-glitch/Knowledge-Graph-Based-Multimodal-E-Learning-Recommender-System.git

2. Install and set up Neo4j:

   • Download and install Neo4j from Neo4j Downloads.
   • Set up the database with your dataset.

Team Members

• Aarush Gupta
• Saksham Purohit
• Shivaprasad Arunkumar Farale

Future Work

• Integrating real-time user interaction features.
• Enhancing the hybrid recommendation models with deep learning techniques.
• Improving graph visualization for better insights.

References

• Neo4j Documentation: https://neo4j.com/docs/
• Louvain Algorithm: Community Detection in Graphs
• Python Libraries: Pandas, Matplotlib, Scikit-learn