Weakly Supervised Cross-platform Teenager Detection with Adversarial BERT
This addresses the challenge of protecting teenagers from negative influences on social media, though it is incremental as it builds on existing transfer learning and adversarial methods.
The paper tackles the problem of detecting teenage users across social media platforms with limited labeled data by proposing a cross-platform framework using Adversarial BERT, which significantly improves over baseline models in experiments on four datasets.
Teenager detection is an important case of the age detection task in social media, which aims to detect teenage users to protect them from negative influences. The teenager detection task suffers from the scarcity of labelled data, which exacerbates the ability to perform well across social media platforms. To further research in teenager detection in settings where no labelled data is available for a platform, we propose a novel cross-platform framework based on Adversarial BERT. Our framework can operate with a limited amount of labelled instances from the source platform and with no labelled data from the target platform, transferring knowledge from the source to the target social media. We experiment on four publicly available datasets, obtaining results demonstrating that our framework can significantly improve over competitive baseline models on the cross-platform teenager detection task.