Well-being, Emotion, Network: Geo-spatial Trackers of Unobservable Human Attributes on Social Media

Date: 

Wednesday, March 30, 2022, 3:30pm to 4:30pm

Location: 

Virtual, see Zoom link below

Presentation by Yichun Fan   

View the slides and the recording of this presentation.

Abstract
Human behaviors and performances are largely determined by unobservable human attributes, which hinder the progress of social and behavioral science. The proliferation of geotagged social media data provides new opportunities to develop instant metrics for these important yet unobservable attributes and to unveil their relationship with local policies/behaviors. I will introduce my past and ongoing research at the MIT Sustainable Urbanization Lab that relies on the cutting-edge natural language processing and econometric techniques to explore the power of large-scale geotagged social media data (Twitter and Weibo) in: (1) Quantifying subjective well-being; (2) Detecting emotional states; (3) Measuring information network across space. Beyond developing real-time metrics, we apply causal inference techniques to carefully examine the determinants and real-world behavioral implications of each human attribute in the context of climate change or global pandemic. I will demonstrate in our applications how geotagged social media data can provide valuable complements to traditional data to support evidence-based decision-making.

Speaker Bio

Yichun Fan is a PhD student at MIT Department of Urban Studies and Planning and a graduate researcher at MIT Sustainable Urbanization Lab. Yichun’s research examines human responses to the leading environmental and health threats in cities to support more effective and equitable urban sustainability policies. She is particularly interested in applying big data and computational methods with causal inference techniques to enhance our understanding of dynamic human behaviors. She received her Master in City Planning from MIT and Bachelor in Environmental Engineering from Tsinghua University.

Join Zoom meeting:

https://harvard.zoom.us/j/94415974099?pwd=WFRER0phcmVGMlU5Ry8vTTEwS2puZz09

Password: 614968

Join by telephone: +1 929 436 2866.  International numbers available: https://harvard.zoom.us/u/acWK12yiQ9