I use data science, NLP, social network analysis, machine learning, and LLM-based methods to study social work, public health, and substance-use data. This track focuses on using computational methods to find patterns, generate evidence, and produce new insights from complex social data.
I study how AI and digital systems become part of social welfare practice, from secure data pipelines to theory, AI-assisted tools, and human oversight.
Foundation
Infrastructure and Validation
Secure data preparation, de-identification, evaluation design, data protection, and production pipelines for AI systems in social service settings.
Theory
Socio-technical Systems Theory
Using socio-technical systems theory to study how workers, agencies, policies, and technical systems shape one another in social welfare practice.
Tools
Welfare Tools
AI-assisted tools for documentation review, policy alignment, caseworker support, and quality assurance. This includes my child welfare policy-alignment RAG system for CPS investigation notes.
Oversight
Human-Centered Oversight
Studying how practitioners and supervisors review, question, evaluate, benchmark, and govern AI outputs in high-stakes public systems. This includes human-in-the-loop and human-on-the-loop oversight, as well as evaluation protocols that test whether AI systems are accurate, useful, and safe for social welfare practice.
Searching the Literature by Meaning: Benchmarking Free and Commercial Word Embedding Models on Social Work Text
Applying statistical machine learning methods to analyze differences in the severity level of COVID-19 among countries
Application of social network analysis of COVID-19 related tweets mentioning cannabis and opioids to gain insights for drug abuse research