Movie Recommender System

🎬A Streamlit-based web application that recommends movies based on user preference. The system uses machine learning to calculate similarity between movies and fetches real-time posters using the TMDB API.Getty Images Explore 🚀 FeaturesSearch & Select: Choose from thousands of movies in the database.Similarity Engine: Uses Cosine Similarity to find the most relevant matches.Dynamic Posters: Fetches high-quality movie posters via The Movie Database (TMDB) API.Responsive UI: A clean, 5-column layout to display recommendations.🛠️ Tech StackFrontend: StreamlitLanguage: Python 3.xLibraries: Pandas, Scikit-learn, Pickle, RequestsAPI: TMDB API 📂Project StructurePlaintext├── model/ │ ├── movie_list.pkl # Processed movie DataFrame │ └── similarity.pkl # Precomputed Similarity Matrix ├── app.py # Main Streamlit application ├── requirements.txt # List of dependencies └── README.md ⚙️ Installation & SetupClone the Repository:Bashgit clone https://github.com/your-username/movie-recommender.git cd movie-recommender Install Dependencies:Bashpip install -r requirements.txt Data Preparation:Ensure you have your .pkl files inside a folder named model. (These are generated from your Jupyter Notebook/Jupyter Lab environment after calculating the similarity matrix).Run the App:Bashstreamlit run app.py 🧩 How It WorksVectorization: The system converts movie tags (genres, keywords, cast, etc.) into vectors.Similarity Score: It calculates the distance between these vectors using Cosine Similarity:$$\text{similarity} = \cos(\theta) = \frac{\mathbf{A} \cdot \mathbf{B}}{|\mathbf{A}| |\mathbf{B}|}$$Fetching Results: When you select a movie, the app finds the 5 movies with the highest similarity scores and displays their titles and posters.