CRC September Colloquium Series: An AI-Assisted Methodology for Quantifying Measurement Error of Social Media Text Posts

Michael G. Elasmar, Ph.D., Associate Professor, Media Science: Can we use AI to help us extract media users’ thoughts and feelings from the texts they post? This presentation proposes and provides a proof of concept for an AI-assisted methodology that aims to quantify validity and reliability in the context of social media text expressions. It demonstrates that there is a lot more to extract from text than sentiment and shows that the traditional practice of analyzing whole text isn’t the best approach.