word cloud from paper abstracts

Much of my research follows a theme that I call user-oriented natural language processing: breaking down lengthy, complex texts that computing users are obligated to read, and automatically finding the salient details that affect them most. This theme brings together natural language processing (NLP), data science, artificial intelligence, digital libraries, privacy, usability, and linguistics.

My recent work with the Usable Privacy Policy Project has used a combination of crowdsourcing, NLP, and machine learning to extract key details from websites' and mobile apps' privacy policies. I am also interested in privacy in online social networks, and within that topic area I have published on location sharing and post deletion.

A line of research that I call "the metalanguage project" is also part of my recent work. This is an effort to automatically identify instances of metalanguage (i.e., statements about language use or production) in text and extract from them direct, precise information about artifacts of communication.

I will provide a more comprehensive research statement upon request. You may also browse my publications.