Supporting Massive Online Learning Communities with Automatic Text Generation

• In Event: Artificial Intelligence in Educational Contexts

Short Description

The purpose of this work is to explore the use of deep learning algorithms for automatic text generation in providing emotional and community support for a massive online learning community, Scratch. We employed state-of-art deep learning language models GPT-2 and Recurrent Neural Network to test on a date set of almost two million comments. Both automatic readability test and human evaluations are conducted to investigate the supporting performance.

Authors

• Contributor: Hanxiang Du, University of Florida

• Presenter: Wanli Xing, University of Florida

• Contributor: Bo Pei, University of Florida