Qualitative coding and evaluation with AI language models (GPT-3, etc.)
Worked on coding and visualizing qualitative data with AI language models for conference presentation and research. Used GPT-3 and similar models to generate, annotate, and evaluate themes from open-ended text responses. Collaborated with team members to compare inductive coding between humans and large language models. • Applied prompt engineering and generative labeling methods to research on adolescent deconversion themes. • Evaluated alignment and accuracy of AI-generated thematic summaries. • Annotated and reviewed qualitative datasets for accuracy and nuance. • Contributed findings to research conferences on psychology and AI.