Stevie Chancellor builds and critically examines human-centered algorithms for high-risk, dangerous health behaviors in online communities. I use digital trace data from millions of interactions on social media to understand and identify high-risk behaviors with machine learning and computational linguistics, combined with interdisciplinary insights from clinical psychology critical data studies.
Along the way, she explores how to conduct human-centered machine learning, an approach that deliberately refocuses technological design and implementation on the needs of humans, communities, and stakeholders. This includes tensions around rigor and robustness, construct validity, platform governance, and ethical issues within this research agenda. Deeply cares about doing right by people and communities, and (recently) have been thinking about how to develop more ethical and compassionate research practices in data-driven approaches.