Movie Recomend

🎬 Movie Recommendation System Using KNN

🔍 Overview

This project is a movie recommendation system built using the K-Nearest Neighbors (KNN) algorithm. The system suggests movies similar to a given movie based on genres and average ratings, leveraging the popular MovieLens dataset.

---

📂 GitHub Repository

🔗 Link to GitHub Repo

🚀 Features

✅ Dataset Merging – Combines movie metadata and user ratings for unified processing.
✅ Genre Encoding – Encodes multi-genre data for better analysis.
✅ Aggregated Ratings – Uses average movie ratings as a key feature.
✅ KNN Implementation – Leverages cosine similarity to recommend similar movies.

---

🔄 Project Workflow

1️⃣ Data Loading

The project uses two datasets:

• Movies Metadata – Contains details like movieId, title, and genres.
• User Ratings – Includes user preferences in the form of ratings.

2️⃣ Data Merging

Datasets are merged on the movieId column to align movie details with their respective ratings.

3️⃣ Feature Engineering

• Multi-genre encoding using MultiLabelBinarizer, converting genres into binary features.
• This allows the model to compare movies based on genre similarity.

4️⃣ Aggregation

For each movie, the following features are calculated:

• Average Rating – Summarizes user feedback.
• Genre Binary Features – Indicates the presence of specific genres.

5️⃣ Recommendation Model

• The KNN algorithm is used to find movies similar to a selected one.
• Cosine similarity measures similarity between movies based on their features.

---

🛠️ Technologies Used

🔹 Python – Programming language for implementation.
🔹 Pandas – Data processing and aggregation.
🔹 Scikit-learn – Encoding genres and implementing KNN.

---

Why KNN?

✔️ Simple and effective – Works well for similarity-based recommendations.
✔️ No assumption of data distribution – Unlike some parametric models.
✔️ Cosine similarity – Ideal for comparing high-dimensional data like movie genres.

---