From None to Severe: Predicting Severity in Movie Scripts

TL;DR

This paper presents a novel approach using a siamese network-based multitask framework to predict the severity of age-restricted content in movie scripts, improving interpretability and outperforming previous models.

Contribution

Introduces a new task for predicting severity in movie scripts with a siamese network multitask model that enhances interpretability and achieves better performance.

Findings

Outperforms previous state-of-the-art models

Provides interpretable predictions

Uses a new publicly available dataset

Abstract

In this paper, we introduce the task of predicting severity of age-restricted aspects of movie content based solely on the dialogue script. We first investigate categorizing the ordinal severity of movies on 5 aspects: Sex, Violence, Profanity, Substance consumption, and Frightening scenes. The problem is handled using a siamese network-based multitask framework which concurrently improves the interpretability of the predictions. The experimental results show that our method outperforms the previous state-of-the-art model and provides useful information to interpret model predictions. The proposed dataset and source code are publicly available at our GitHub repository.