Movie Genre Prediction

This project aims to predict the genre of a movie based on its plot summary. Leveraging Natural Language Processing (NLP) and machine learning techniques, the system processes textual data and applies various models to achieve accurate predictions.

Directory Structure:

Directory structure:
└── 04bhavyaa-movie-genre-prediction/
    ├── app.py
    ├── artifacts.dvc
    ├── data/
    │   ├── train_data.txt
    │   ├── test_data_solution.txt
    │   ├── description.txt
    │   └── test_data.txt
    ├── requirements.txt
    ├── genre-classification.ipynb
    └── README.md

Tech Stack:

• Libraries: Scikit-learn, Pandas, NumPy, Matplotlib, Seaborn, NLTK
• Models: Logistic Regression, Linear SVC, Random Forest, Naive Bayes
• Best Performing Model: Ensemble of Linear SVC and Logistic Regression with an accuracy of 58.88% on the validation set.

Key Takeaways:

• Ensemble models outperformed individual models due to combined decision-making.
• Fine-tuning hyperparameters significantly improved accuracy for individual models.
• Preprocessing steps like lemmatization and scaling were essential for handling text data.

App Gallery: