--- license: apache-2.0 datasets: - pratikkalamkar/Balanced_Hollywood_Movies_Scripts_Age_Rating_Dataset_250 language: - en base_model: - google-bert/bert-base-uncased pipeline_tag: text-classification library_name: transformers metrics: - accuracy - precision - recall - f1 - mae tags: - movie-certification - movie-scripts - screenplay - mpaa-rating - age-rating - ordinal-classification - hierarchical-bert - bert - attention-pooling - explainable-ai - english - ieee-gcat-2025 --- # Hierarchical BERT with Attention Pooling for MPAA Age Rating Prediction This repository contains a **Hierarchical BERT** model with **attention pooling** and **ordinal learning** trained on the **Balanced Hollywood Movie Scripts Age Rating Dataset (250 scripts)** for automated **MPAA age rating prediction** from full-length English movie scripts. The model uses **google-bert/bert-base-uncased** as the base transformer and was developed as part of the following publication: > **Hierarchical Ordinal Framework for Automated Movie Censorship Using Full-Length Scripts** ## Dataset - **Balanced Hollywood Movie Scripts Age Rating Dataset (250)** - 250 English movie scripts - Balanced across five MPAA age rating categories: - G - PG - PG-13 - R - NC-17 - 50 scripts per class Dataset: https://huggingface.co/datasets/pratikkalamkar/Balanced_Hollywood_Movies_Scripts_Age_Rating_Dataset_250 ## Task Predict the appropriate MPAA age rating for a complete English movie script using a **Hierarchical BERT** architecture with **attention pooling** and **ordinal learning**. ## Citation If you use this model in your research, please cite: ```bibtex @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," 2025 IEEE 6th Global Conference for Advancement in Technology (GCAT), Bangalore, India, 2025, pp. 1-7, doi: 10.1109/GCAT66372.2025.11368510. ## License Apache License 2.0