Brian Keith N.

Email: briankeithn [AT] vt [DOT] edu

About Me

I am now working at Universidad Católica del Norte as an Assistant Professor. This site is outdated! Check out my new website instead. 

I came to Virginia Tech to work towards a Ph.D. in Computer Science under the Fulbright Faculty Development program, sponsored by Universidad Católica del Norte. In addition to this support, I am also a Research Assistant at Virginia Tech and I have received a scholarship from the Chilean Government ("Becas Chile"). As part of the Fulbright Faculty Development program, once I return to Chile, I'll work as an Assistant Professor in my sponsoring institution. 

My research interests lie in Natural Language Processing, Visual Analytics and Social Computing. As of right now I'm researching ways to represent and visualization information narratives with my advisors Dr. Chris North and Dr. Tanushree Mitra.

Education

I started my PhD in Computer Science at Virginia Tech on Fall 2019. I'm currently working towards completing my breadth requirements.

All my previous degrees and professional qualifications were obtained at Universidad Católica del Norte, Antofagasta, Chile.

Publication Highlights

B. Keith & T. Mitra - Accepted in CSCW 2020

Narratives are fundamental to our perception of the world. By combining the theory of narrative representations with the data from modern online systems, we make three key contributions: a theory-driven computational representation of narratives, a novel extraction algorithm to obtain these representations from data, and an evaluation of our approach.

Evaluating the Inverted Pyramid Structure through Automatic 5W1H Extraction and Summarization [PDF]

B. Keith, M. Horning & T. Mitra - C+J 2020, USA.

The inverted pyramid structure is a cornerstone of journalism associated with neutrality and objectivity. Can we determine how well an article follows this structure? We propose a quantitative approach using summarization and 5W1H extraction to do this in an Associated Press news articles data set. 

Sentiment Analysis and Opinion Mining Applied to Scientific Paper Reviews [PDF]

B. Keith, E. Fuentes & C. Meneses - Intelligent Data Analysis, volume 23(1), 2019.

Peer-reviewed conferences face several challenges during the review process. In particular, how consistent are the evaluations given by reviewers with the contents of the review itself? Using sentiment analysis hybrid methods, we determine the sentiment orientation of paper reviews, thus presenting the first step towards answering this question.