🎬 Netflix Content Classification & Recommendation System

An AI-powered system that classifies Netflix titles (Movie vs. TV Show) and recommends similar content using both TF-IDF and Transformer-based semantic models.
Built with Python, scikit-learn, Sentence Transformers, and Gradio for an interactive UI.

---

📌 Project Overview

OTT platforms like Netflix have thousands of titles, making it challenging for users to discover relevant content.
This project aims to:

• Automatically classify Netflix titles as Movie or TV Show using ML models.
• Provide intelligent recommendations using both keyword-based and semantic similarity.
• Deliver an interactive UI for end users and stakeholders.

---

🚀 Features

• Classification Module
  • TF-IDF feature extraction.
  • Logistic Regression, SVM, and Naive Bayes classifiers.
  • Tuned with GridSearchCV for optimal performance.
• Recommendation Engine
  • TF-IDF based recommender: Keyword-level similarity.
  • Semantic recommender: Sentence-BERT (all-MiniLM-L6-v2) embeddings for context-aware recommendations.
• Visualization
  • Distribution of Movies vs TV Shows.
  • Release year trends.
  • Genre and country analytics.
  • Confusion matrices and classification reports.
• Interactive UI
  • Built with Gradio.
  • Two Tabs:
    1. Classification (predict Movie/TV Show).
    2. Recommendations (TF-IDF & Semantic suggestions).

---

🛠️ Tech Stack

• Languages & Libraries: Python, Pandas, NumPy, Matplotlib, Seaborn
• ML/NLP: scikit-learn, Sentence-Transformers
• Deployment: Gradio
• Dataset: Netflix Titles Dataset (Kaggle)

---