Toward the pursuit of educational and social equity with data science and AI, I approach three strands of research:
- Leverage data science and AI as a microscope to generate fine-grained understanding of inequality;
- Investigate the potential of data science and AI as a remedy for existing inequalities;
- Understand the inherent equitability of data science and AI techniques.
Empirically, I work with novel forms of “big data” generated in real-world educational contexts and focus on under-resourced institutions and marginalized populations.
Current projects
New projects are always on the way!
Longitudinal modeling of educational inequality
We investigate modeling strategies that convert longitudinal, unstructured data (e.g., digital traces, curricular content) into a rigorous understanding of how educational inequality accumulates through day-to-day experience.
Algorithmic fairness in transfer learning
We examine sources, measures, and mitigation strategies of algorithmic bias in cross-context transfer learning to improve the robustness and generalizability of educational models especially in under-resourced environments.
Digital divides in the age of AI
We leverage new data sources and analytical tools to track new forms of digital inequalities and unintended social consequences incurred by the rapid development of AI.