Anime Recommendation System 🎌🤖

A personalized anime recommendation system using neural networks and content-based filtering, trained on the Kaggle "Manga, Manhwa, and Manhua Dataset".

---

🚀 Project Overview

This project leverages deep learning to predict anime a user might enjoy based on their historical preferences. By transforming anime tags and user interactions into dense vectors, we train a neural network that learns the hidden relationships between user tastes and anime features.

---

🔍 Features

• 📚 Content-based filtering using tag metadata
• 🧠 Neural network recommender (feedforward binary classifier)
• 🎯 User and anime embedding using averaged tag vectors
• 📊 Evaluation metrics: Accuracy, ROC-AUC, Precision@K
• 🌐 Ready for frontend integration via Streamlit

---

🗂️ Project Structure

anime-recommender/
├── anime_recommender.ipynb # Cosine similarity + content filtering
├── NN.ipynb # Neural network training
├── NN_anshuman.ipynb # Refined neural model
├── user_ready_final.ipynb # User profile generation
├── user_relevant_infoextraction.ipynb # Metadata + tag handling
├── README.md # This file
└── requirements.txt # Dependencies

---

🧪 Model Architecture

• Input: Concatenated user and anime tag vectors
• Hidden Layers: Dense layers with ReLU + Dropout
• Output: Sigmoid activation (like vs. not-like)
• Optimizer: Adam
• Loss: Binary Cross-Entropy

---

📦 Setup

# Create environment
python -m venv anime-env
source anime-env/bin/activate

# Install dependencies
pip install -r requirements.txt