Wenyu Zeng

Wenyu Zeng

she/her

PhD student in Biomedical & Health Informatics at the University of Washington. My research focuses on building scalable computational pipelines that integrate spatial transcriptomics and single-cell sequencing data to study glioblastoma heterogeneity. I’m passionate about combining biomedical data science and translational research to better understand tumor biology and identify therapeutic targets. Beyond research, I enjoy data visualization, teaching, and exploring new ways to make complex data accessible and meaningful.

Learning from Elders: Making an LLM-powered Chatbot for Retirement Communities more Accessible through User-centered Design

Published in arXiv preprint arXiv:2504.08985, 2025

We designed an LLM-powered chatbot prototype tailored for retirement community residents to bridge gaps in technology and eHealth literacy.
Using a human-centered design process, we conducted interviews, built personas, and refined usability through pilot testing.
The final chatbot—built with GPT-3.5 Turbo and Streamlit—supports adjustable text size, theme control, and personalized follow-ups.
Results from a pilot trial showed high satisfaction and ease of use, emphasizing the potential of accessible, AI-driven interfaces to empower older adults.

Recommended citation: Li, L. X., Chung, R., Chen, F., Zeng, W., Jeon, Y., & Zaslavsky, O. (2025). "Learning from Elders: Making an LLM-powered Chatbot for Retirement Communities more Accessible through User-centered Design." arXiv preprint arXiv:2504.08985. https://doi.org/10.48550/arXiv.2504.08985
Download Paper