Haofei Xu

Haofei Xu

Computational Social Science Researcher

Hi there! I am Haofei Xu, a Ph.D. student in Computational and Data Science at Washington University in St. Louis, where I began my doctoral studies in August 2025. I am passionate about leveraging computational methods to understand complex social and political phenomena, with a focus on natural language processing, machine learning, and large-scale data analysis.

My research centers on applying advanced computational techniques to analyze political discourse and social behavior. I have extensive experience in developing scalable data pipelines, implementing state-of-the-art NLP models, and conducting large-scale text analysis. My work spans from sentiment analysis of political communications to topic modeling of public discourse in local governance.

I bring a unique interdisciplinary perspective to my research, combining my background in social psychology with expertise in computer science, statistics, and data engineering. This diverse foundation allows me to bridge the gap between technical innovation and social science applications, developing computational solutions that provide meaningful insights into human behavior and democratic processes.

Research Interests

My research focuses on:

  • Natural Language Processing and Text Mining
  • Political Discourse Analysis and Computational Social Science
  • Machine Learning for Social Good
  • Large-Scale Data Processing and Analysis
  • Topic Modeling and Sentiment Analysis
  • Democratic Participation and Public Engagement

Education

  • Ph.D. in Computational and Data Science, Washington University in St. Louis, 2030 (expected)
  • M.S. in Information, University of Michigan, 2024
  • B.S. in Social Psychology, with minors in Computer Science and Statistics, University of Illinois at Urbana-Champaign, 2022

Recent Research Highlights

My current research focuses on computational analysis of political discourse and democratic participation. Key projects include:

  • Large-Scale Political Discourse Analysis: Leading a project analyzing public comments from 500+ Michigan municipalities, implementing scalable ETL pipelines on AWS and advanced NLP techniques including BERTopic for topic modeling.

  • Election Sentiment Analysis: Developed sentiment analysis models for 2024 US Election tweets using transformer models (DistilBERT), achieving 98.54% accuracy on 1M+ tweet dataset.

  • Accessibility Research: Conducted comparative analysis of web accessibility standards between US and Chinese universities, combining computational analysis with field studies.

Publications and Research

You can find my publications on my Google Scholar profile. My recent work includes publications on political discourse analysis and public engagement in local governance.

For a complete overview of my academic background and experience, please see my CV.

News

  • August 2025 - Started Ph.D. program in Computational and Data Science at Washington University in St. Louis
  • April 2024 - Presented research at Midwest Speech and Language Days 2024