RESEARCH COLLOQUIUM ON COMPUTATIONAL SOCIAL SCIENCE/DATA SCIENCES – Xiaoyi Yuan – Understanding Urban Places Through Textual Volunteered Geographic Information and Machine Learning

When:
March 20, 2020 @ 3:00 pm – 4:00 pm
2020-03-20T15:00:00-04:00
2020-03-20T16:00:00-04:00
Where:
Center for Social Complexity, 3rd Floor Research Hall
Cost:
Free
Contact:
Karen Underwood
7039939298

Research Colloquium on Computational Social Science/Data Science

Xiaoyi Yuan
CSS PhD Candidate

Understanding Urban Places Through Textual Volunteered Geographic Information and Machine Learning

Friday, March 20, 3:00 p.m.
Center for Social Complexity
3rd Floor Research Hall

All are welcome to attend.

Abstract:
Leveraging textual Volunteered Geographic Information (VGI), such as geolocated Tweets and online reviews allow us to discover knowledge of people’s experiences and perceptions of places on a large scale. This Friday Seminar includes two recent studies that analyze textual VGI through machine learning, one of which demonstrates how to use deep learning (Convolutional Neural Networks) and fine grained aspect-based sentiment analysis on Yelp reviews to understand restaurant culture in various urban areas. The second study showcases how we can combine topic modeling and network analysis to discover complex relationships between places in Manhattan, NY using Twitter and TripAdvisor reviews. In addition, it shows that different data sources pose opportunities and challenges for studying urban places.
Bio:
Xiaoyi Yuan is a PhD candidate in the Computational Social Science program at George Mason University. She works primarily with her advisor Andrew Crooks on utilizing machine learning to analyze online crowdsourced geo-enriched textual data for understanding urban place perceptions. Prior to her PhD, she received her Masters degree in Communication, Culture, and Technology at Georgetown University, focusing on computational techniques in analyzing framing of news articles and discovering online communication patterns.