Movie Recommendation System using Content-Based Filtering

🎯 Project Overview

This project implements a content-based movie recommendation system using metadata such as genres, keywords, cast, crew, and overview to suggest similar movies.

🧰 Features & Workflow

• Extracts and processes key movie features: genres, keywords, overview, cast, and crew
• Cleans and normalizes data (e.g., removing spaces in names like SamMendes)
• Constructs a new tags column by combining textual data from multiple features
• Converts text into numeric vectors using Bag of Words (BoW)
• Computes similarity between movies using cosine similarity
• Returns the top N recommended movies for a selected title

🧠 Techniques Used

• Natural Language Processing (NLP)
• CountVectorizer for feature extraction
• Cosine similarity for recommendation logic
• Data preprocessing using Pandas and NumPy

▶️ How to Use

1. Clone the repository:

   git clone https://github.com/rohitsahayy/movie-recommender-system.git
   cd movie-recommender-system

2. Open main.ipynb in Jupyter or Google Colab.

3. Run all cells and use the recommend() function to get movie recommendations.

📦 Dependencies

• Python 3.7+
• pandas
• numpy
• scikit-learn
• nltk

Install them via:

pip install pandas numpy scikit-learn nltk

📌 Example

recommend('Avatar')

Returns a list of movies similar to Avatar based on metadata similarity.