To Prospective Ph.D. Students, Post-docs, and Visitors
I'm interested in doing research in natural language processing (NLP),
where I design machine learning algorithms and computational models
(e.g., neural networks, stochastic algorithms, probabilistic models, etc)
for tasks on abstractive text summarization, language generation,
argumentation mining, discourse analysis, and dialog analysis.
I'm also interested in applying NLP and machine learning techniques
for interdisciplinary subjects, e.g. computational social science.
Specifically, our group has the following ongoing research projects:
We tackle the challenge of extracting key information from large amounts
of and long documents. We aim to generate concise and informative summaries
for different types of texts, ranging from news articles in traditional media,
to socially-generated content in popular social media (e.g. comments, tweets, or blogs),
and to government meetings (e.g. Federal Reserve board meetings).
We are interested in natural language generation (NLG) techniques, where the systems have controllability over the content and discourse structure of the outputs.
For instance, we have worked on persuasive argument generation, which requires understanding of many aspects of high quality arguments,
including discourse, semantic, and linguistic style.
Related directions include reliable and trustworthy text generation, complex question generation, and bias mitigation in NLG systems.
Arguments play an important role for decision-making processes and persuasion.
We are interested in understanding how people argue with and influence others,
as well as form their own opinions on topics of interest.
Especially, we want to discover linguistic patterns that reflect these processes,
and use them for social interaction analysis and prediction.
NLP models for narrative structure, discourse, and semantic analysis are investigated for this project.
Media Bias Analysis:
News media play a vast role not just in supplying information, but in selecting, crafting, and biasing that information to achieve both nonpartisan and partisan goals. We aim to automate media bias detection from news articles, and quantify and further highlight biased content in order to promote the transparency of news production as well as enhance readers’ awareness of media bias. Discourse analysis and opinion mining algorithms are designed to identify and categorize different types of bias, which in turn facilitate the understanding of the prevalence of bias in content produced by media with different ideological leanings. Potentially applications include misinformation detection and characterization, multi-perspective news summarization, etc.
If any of these sounds interesting to you, please check out more about
my research and papers at this website.
If you find your research agenda might be aligned with mine
and are interested in working with me, please fill in this
external contact form
and (for Ph.D. applicants) apply to Computer Science and Engineering
at University of Michigan and mention my name in your application.