MovieSense: Movie Recommender System

Project description

MovieSense is a Recommender System that uses Natural Language Processing techniques to provide users with similar movies to a movie and relevant information about the recommendations.This is done by creating embeddings of the movie features from the dataset, transformation of data to vectors and computing the similarities of movies. This project aims to tackle the problem of choice overload.

Features

1. NLP Techniques and ML Models:
   • Vectorization: Term Frequency-Inverse Document Frequency vectorizer
2. Mathematical Technique:
   • Similarity: Cosine Similarity
3. Framework/Libraries:
   • Gradio: web application user interface
   • scikit-learn
   • Transformers (Hugging Face)

Table of Contents

Getting Started

The contents of the repository may be cloned or downloaded and the demo_app notebook run using a jupyter notebook or from colab. Repository structure should be maintained.

Installation

1. Clone the repository: git clone https://github.com/AISaturdaysLagos/Cohort8-Sankara.git
2. Navigate to the project directory: cd Cohort8-Sankara
3. Install required packages: pip install -r requirements.txt
4. Run the main script: python app.py
5. Click the corresponding link from terminal to view and use app

Web Application Deployment Link

• https://huggingface.co/spaces/CharlyiE/Team_Sankara

Acknowledgments

• David Onyeali(Mentor)
• AI Sturday Lagos team
• scikit-learn Library
• Team members

Contact

• David: [email protected]
• Charles: [email protected]
• Ehis: [email protected]
• Misi: [email protected]