movie-recommender-system

This repository contains the implementation and analysis of a movie recommendation system. The system utilizes the MovieLens dataset to implement and evaluate multiple recommendation models, such as content-based filtering, Singular Value Decomposition (SVD), Bayesian Parameter Models, and Autoencoders. The project focuses on addressing challenges like data sparsity, scalability, and cold-start problems while maximizing recommendation accuracy.

Description

The system utilizes the MovieLens Dataset containing:

• Metadata for over 45,000 movies.
• 26 million user ratings from more than 270,000 users.

---

Models Implemented

1. Content-Based Filtering

2. Singular Value Decomposition (SVD)

3. Bayesian Parameter Model

4. Autoencoders

---

How to Run the Code

1. Clone this repository:

   git clone https://github.com/your-repo-url
   cd movie-recommendation-system

2. Install dependencies:

   pip install -r requirements.txt

3. Run the Jupyter Notebook:

   jupyter notebook movie_recommender.ipynb

4. Follow the notebook to train and evaluate different models.

---

Evaluation Metrics

• Root Mean Squared Error (RMSE): Measures prediction accuracy.
• Mean Absolute Error (MAE): Focuses on the average magnitude of prediction errors.
• Precision@K: Proportion of relevant recommendations in the top-K items.
• Recall@K: Proportion of relevant items retrieved in the top-K items.