🎬 Movie Rating Prediction using Machine Learning

This project predicts IMDb movie ratings based on key features such as release year, duration, and number of votes. It uses a regression model built with Python, scikit-learn, and Pandas, and follows a complete machine learning pipeline from data preprocessing to model deployment.

---

📊 Project Overview

• Dataset: IMDb Movies India.csv (from Kaggle)
• Goal: Predict movie ratings using features like Year, Duration, and Votes
• Model Used: SGDRegressor (Stochastic Gradient Descent)
• Libraries: pandas, numpy, matplotlib, seaborn, scikit-learn

---

🔧 Steps Performed

1. Data Preprocessing

• Handled null values and cleaned feature columns (e.g., removed symbols from strings)
• Converted datatypes (Year, Votes, Duration)
• Removed less relevant columns (e.g., Genre, Actors)

2. Exploratory Data Analysis (EDA)

• Used histograms, boxplots, and scatterplots to understand feature distribution
• Plotted a heatmap for correlation analysis

3. Feature Engineering

• Selected numerical features: Year, Duration, Votes as inputs (X)
• Used Rating as the target output (y)

4. Model Building

• Created a pipeline with:
  • StandardScaler for feature scaling
  • SGDRegressor for regression
• Split data into training and testing sets

5. Model Evaluation

• Evaluated using:
  • Mean Absolute Error (MAE)
  • Mean Squared Error (MSE)
  • R² Score

6. Model Deployment

• Accepted user input for new movie data
• Predicted rating using trained pipeline

---

✅ Example Prediction

new_input = pd.DataFrame({
    'Year': [2023],
    'Duration': [120],
    'Votes': [10000],
})
predicted_rating = pipeline.predict(new_input)
print("Predicted Rating:", predicted_rating)