Detecting Content Rating Violations in Android Applications: A Vision-Language Approach

TL;DR

This paper introduces a vision-language model to automatically predict and detect content rating violations in Android games, outperforming previous models and revealing potential non-compliance cases in the Play Store.

Contribution

The study presents a novel multi-modal approach using vision and language data to identify content rating violations, improving accuracy over existing models.

Findings

Achieved ~6% higher accuracy than the state-of-the-art CLIP-based model.

Detected over 70 potential content rating violations in real-world apps.

Found that apps flagged by the classifier had a higher removal rate from the Play Store.

Abstract

Despite regulatory efforts to establish reliable content-rating guidelines for mobile apps, the process of assigning content ratings in the Google Play Store remains self-regulated by the app developers. There is no straightforward method of verifying developer-assigned content ratings manually due to the overwhelming scale or automatically due to the challenging problem of interpreting textual and visual data and correlating them with content ratings. We propose and evaluate a visionlanguage approach to predict the content ratings of mobile game applications and detect content rating violations, using a dataset of metadata of popular Android games. Our method achieves ~6% better relative accuracy compared to the state-of-the-art CLIP-fine-tuned model in a multi-modal setting. Applying our classifier in the wild, we detected more than 70 possible cases of content rating violations, including nine instances with the 'Teacher Approved' badge. Additionally, our findings indicate that 34.5% of the apps identified by our classifier as violating content ratings were removed from the Play Store. In contrast, the removal rate for correctly classified apps was only 27%. This discrepancy highlights the practical effectiveness of our classifier in identifying apps that are likely to be removed based on user complaints.