Social science deserves better AI tools - models designed on true understanding, not just prediction. For too long, computational social scientists struggled with black-box models prioritising opaque precision over interpretation. And we’re leading that change. Our goal is to create AI tools that embrace what makes social science so valuable, which is the pursuit of true understanding. That is why our research focuses on integrating AI models with core social science principles, enhancing interpretability, transparency, and accountability - areas where social science and modern AI diverge.
August 2023
Fusing AI linguistics with configurable logic for interpretable sentiment analysis
Learn moreTailor sentiment analysis to your needs with SentiBank, the largest public repository of sentiment dictionaries.
Browse sentibank