Analyzing Human and Social Dynamics Through Social Sensing
PI: Xi Gong (Biobehavioral Health, ICDS)
The level of effort appropriate for the proposed project: 1 semester at 8 hours a week.
Plan for funding tuition (for graduate students): My unit is currently unable to cover tuition. (There is a possibility that tuition support may be available if the student is a BBH graduate student, but this is not guaranteed.)
A brief (up to 1 page) description of the proposed project:
Social media data offers unprecedented opportunities for human and social research. Compared to traditional data collection methods, such as surveys and interviews, data collected via social sensing (e.g. social media data, cellphone based mobility data) is more cost-effective, geo-referenced, timelier, and available in greater volumes which therefore complement traditional approaches. As a form of spatial-temporal big data, social sensing provides unique insights into spatial-temporal patterns in peoples, places, environments as well as their relationships and social dynamics. My team has innovatively combined social sensing big data with various social science theories and data sources to model and analyze the spatial-temporal aspects of human/social dynamics, such as communications, outreaching, social identities, and public attitudes. While numerous opportunities are available, challenges persist, including sample representativeness, sentiment analysis misclassification, geotag inaccuracies, and rising costs of data collection. This project aims to expand the current study using social sensing for understanding spatial social networks and public perspectives on controversial social topics, also exploring dealing with the challenges inherited in social sensing research.
Planned Activities:
Topic Detection and Tracking: Analyze spatiotemporal patterns in the collected social sensing big data to identify and extract trending topics being discussed.
Sentiment Analysis: Manually annotate study-specific samples and train custom machine learning/AI classifiers to predict sentiment in individual social media posts.
Visualization: Visualize the original data and analytical results through maps and graphs to support interpretation and communication of findings.
Location Spoofing Detection: Explore the development of methods to detect and address location spoofing in social sensing data.
A list of specific areas of computational and/or data science expertise or skills that the current team is particularly interested in recruiting to support the project:
General AI/ML methodology and implementation, Network analysis, Spatial data analysis.
Other Expectations of ICDS Junior Researcher:
Regular availability for meetings (weekly or bi-weekly, times flexible) A list of specific objectives for work supported by this call: We plan to submit at least one manuscript to disseminate the findings. In the medium to long term, we aim to leverage these preliminary results to explore potential links to health outcomes, thereby supporting a future funding proposal to NIH/NIMH and other relevant funding opportunities.
Connection of the project to ICDS’s mission:
We will develop and apply data science and ML/AI methods, and spatiotemporal analysis within the framework of Spatially Integrated Social Science (SISS) to advance the understanding of complex human and social dynamics.
A paragraph summarizing team member’s recent and/or planned engagement with ICDS:
Xi Gong is a co-hire at ICDS. He has regularly participated in and plans to continue engaging in ICDS activities, such as monthly lunches, faculty search interviews, AI Week, and the annual ICDS symposium.