Zhao Wang Ph.D. Assistant Instructional Professor in MACSS The University of Chicago Google Scholar

Bio

• Dr. Wang is an Assistant Instructional Professor in the Masters in Computational Social Science program at The University of Chicago since Fall 2021. Her research is at the intersection of Machine Learning, Natural language Processing, Causal Inference, and Social Network Analysis. She has worked on improving statistical model robustness, exploring computational linguistic models to analyze data from social media platforms and provide insights into social issues, and understanding causal inference in natural language. Examples include identifying spurious correlations in text classifier, improving statistical machine learning model robustness and fairness with causal knowledge, understanding wording effect on human perception, and etc.

Research Interests

• Causal Text Analysis: exploring causes, effects, and causal relationships in natural language.
• Robust models for Text Analysis: developing simple, robust, fair, transparent, and interpretable machine learning models.
• Fair and Trustable AI for social good: applying AI techniques to get insights into social issues (e.g., public health, social bias, fake news, bot detection, cyberbullying, sentiment analysis, human well-being, personality and behavior) and provide practical solutions.

Selected Publications

Enhancing Model Robustness and Fairness with Causality: A Regularization Approach
Zhao Wang, Kai Shu, Aron Culotta.
EMNLP, the First Workshop on Causal Inference and NLP, 2021.
[Paper]

Robustness to Spurious Correlations in Text Classification via Automatically Generated Counterfactuals.
Zhao Wang, Aron Culotta.
In Thirty-Fifth Association for the Advancement of Artificial Intelligence Conference (AAAI), 2021.
[Paper]

Identifying Spurious Correlations for Robust Text Classification.
Zhao Wang, Aron Culotta.
The Conference on Empirical Methods in Natural Language Processing (Findings of EMNLP), 2020.
[Paper] [Code]

When do Words Matter? Understanding the Impact of Lexical Choice on Audience Perception using Individual Treatment Effect Estimation.
Zhao Wang, Aron Culotta.
In Thirty-Third Association for the Advancement of Artificial Intelligence Conference (AAAI), 2019.
[Paper] [Code]

Personality and Behavior in Team-based Online Games.
Zhao Wang, Anna Sapienza, Aron Culotta and Emilio Ferrara.
In IEEE Conference on Games (COG), 2019.
[Paper]

Are Words Commensurate with Actions? Quantifying Commitment to A Cause from Online Public Messaging.
Zhao Wang, Jennifer Cutler, and Aron Culotta.
In IEEE International Conference on Data Mining (ICDM) Workshop on Data science for Human Performance in Social Networks, 2017.
[Paper] [Poster] [Code]

IAfinder: identifying potential implicit assumptions to facilitate validation in medical cyber-physical system.
Zhicheng Fu, Zhao Wang, Chunhui Guo, Zhenyu Zhang, Shangping Ren, Lui Sha.
Design Automation Conference (DAC), 2018.
[Paper]