COMPUTATIONAL SOCIAL SCIENCE SEMINAR – Mining Social Data for Public Behaviors to Improve Absorptive Capacity of Organizations for Decision Support – Hemant Purohit
COMPUTATIONAL SOCIAL SCIENCE FRIDAY SEMINAR
Hemant Purohit, Assistant Professor
Information Sciences and Technology
George Mason University
Mining Social Data for Public Behaviors to Improve Absorptive
Capacity of Organizations for Decision Support
Friday, October 14, 3:00 p.m.
Center for Social Complexity Suite
3rd Floor, Research Hall
ABSTRACT: Social Media or Web 2.0, one source of Big Data, has completely revolutionized information sourcing, management, and processing. The opportunity to understand and exploit such unprecedented level of human interaction data can help public sector organizations improve their processes and decision making, such as for disaster management in building resilient smart cities. The data challenges of large scale volume of public data, velocity of content generation, sparsity of data behaviors, variety of language complexity and diversity of community demographics in this online interaction data present an exciting venue for improving science of public behavior in mediated communication. For instance, mining intentional behaviors of resource seeking on social media helps disaster response organizations better coordinate resources, given the mounting cost of response estimated to be 271 billion USD/yr by 2025.
This talk presents a Web information processing framework with examples to interpret, manage, and integrate unstructured social media data generated by public (citizen sensors) into organizational workflows to improve their absorptive capacity of this unconventional data sources. Via use-cases of disaster response coordination, and gender-based violence policy recommendation, the talk will present experiences and challenges in model public behaviors (e.g., intention, attitude) while tackling data challenges mentioned above. The talk will discuss methods to perform such behavioral computing by fusing knowledge from the Web resources (e.g., Wikipedia, Linked Open Data) and socio-psychological theories of actionable behavior into statistical methods of text mining, and machine learning.
Hemant Purohit is an interdisciplinary, computational social science researcher and an assistant professor of Information Sciences and Technology at GMU. He was one of the ITU Young Innovator 2014 for UN’s ICT agency, for winning a global challenge on Open Source Technologies for Disaster Management, as well as one of the eight international fellows of USAID, Google and ICT4Peace foundation at a key humanitarian technology event of CrisisMappers, ICCM-2013 at UN Nairobi. He has presented several talks, lectures, and tutorials on social computing at prestigious venues including AAAI and SIAM conferences as well as published at and served as reviewer for peer-reviewed conferences and journals including ICWSM, WWW, Journal of CSCW, and ACM TIST. More about Hemant: http://ist.gmu.edu/~hpurohit