Cross-Cultural AI
Studying how language, emotion, and social behavior vary across cultures, platforms, and communities -- from emoji usage to politeness norms to stress expression across the East and West.
We study cultural variation in language, emotion expression, and health behaviors across diverse populations, with the goal of building AI systems that are culturally aware and equitable.
Key areas:
- Cross-cultural emoji usage and digital communication norms
- Politeness, stress, and emotional expression across Eastern and Western cultures
- Culturally-aware NLP for health applications in India, China, and sub-Saharan Africa
- Gender norms in global media, in collaboration with the World Bank Group
- Bias detection and mitigation in health AI models