pratikkalamkar/UnBalanced_Hollywood_Movies_Scripts_Age_Rating_Dataset_1142

Unbalanced Hollywood Movie Scripts Age Rating Dataset

Author: Pratik N. Kalamkar

Description

The Unbalanced Hollywood Movie Scripts Age Rating Dataset is a benchmark dataset containing 1,142 full-length English movie scripts annotated with their official MPAA age ratings. It was created to support research in automated movie censorship, age-rating prediction, content classification, ordinal learning, and long-document Natural Language Processing (NLP).

Unlike balanced benchmark datasets, this collection preserves the natural distribution of MPAA age-rating categories found in publicly available movie scripts, making it suitable for evaluating models under real-world class imbalance.

Dataset Structure

Each movie is provided as an individual UTF-8 encoded text file containing the complete movie script.

Filename Format

[MPAA_Age_Rating]_[Movie_Title]_[Movie_Year].txt

Examples:

G_Toy_Story_1995.txt
PG_Frozen_2013.txt
PG-13_Avatar_2009.txt
R_The_Matrix_1999.txt
NC-17_Showgirls_1995.txt

Each filename includes:

• Official MPAA age rating
• Movie title
• Release year

This naming convention facilitates filtering, benchmarking, and reproducible experimentation.

Dataset Statistics

• Total movie scripts: 1,142
• Language: English
• Format: UTF-8 text files
• Document type: Full-length movie scripts
• Labels: MPAA age ratings

MPAA Age Rating Categories

• G
• PG
• PG-13
• R
• NC-17

Research Motivation

Movie age-rating prediction is an ordinal classification problem because the labels possess a natural ordering based on increasing content restriction.

Most publicly available movie-script datasets are naturally imbalanced. This dataset preserves that real-world distribution, making it valuable for evaluating the robustness of machine learning and deep learning models under realistic conditions.

The dataset is suitable for benchmarking:

• Automated movie censorship systems
• Age-rating prediction models
• Ordinal classification algorithms
• Long-document NLP models
• Machine learning methods
• Deep learning architectures
• Transformer-based models

Applications

• Automated Movie Censorship
• Movie Age Rating Prediction
• Ordinal Learning
• Text Classification
• Long Document Analysis
• Natural Language Processing
• Artificial Intelligence Research

Dataset Construction

The dataset was constructed by collecting publicly available full-length movie scripts from screenplay archives and script repositories, then associating each script with its verified MPAA age rating and release year.

Each script was manually organized, verified, renamed using a standardized filename format, and categorized according to its official MPAA certification.

Citation

If you use this dataset in your research, please cite:

@inproceedings{kalamkar2025hierarchical,
  author    = {Pratik N. Kalamkar and Yogesh K. Sharma},
  title     = {Hierarchical Ordinal Framework for Automated Movie Censorship Using Full-Length Scripts},
  booktitle = {2025 IEEE 6th Global Conference for Advancement in Technology (GCAT)},
  year      = {2025},
  pages     = {1--7},
  doi        = {10.1109/GCAT66372.2025.11368510}
}

or

P. N. Kalamkar and Y. K. Sharma, Hierarchical Ordinal Framework for Automated Movie Censorship Using Full-Length Scripts, Proceedings of the 2025 IEEE 6th Global Conference for Advancement in Technology (GCAT), Bangalore, India, 2025, pp. 1–7. DOI: 10.1109/GCAT66372.2025.11368510

Acknowledgment

This dataset was created to support reproducible research in automated movie censorship and movie age-rating prediction. If you use this dataset in your research, publications, or software, please consider citing the associated paper.