Scalable, Secure Analysis of Social Sciences Data on the Azure Platform

Conference · Mon May 07 00:00:00 EDT 2012

OSTI ID:1332547

Simmhan, Yogesh; Deng, Litao; Kumbhare, Alok; Redekopp, Mark; Prasanna, Viktor

Human activity and interaction data is beginning to be collected at population scales through the pervasiveness of social media and willingness of people to volunteer information. This can allow social science researchers to understand and model human behavior with better accuracy and prediction power. Political and social scientists are starting to correlate such large scale social media datasets with events that impact society as evidence abound of the virtual and physical public spaces intersecting and influencing each other [1,2]. Managers of Cyber Physical Systems such as Smart Power Grid utilities are investigating the impact of consumer behavior on power consumption, and the possibility of influencing the usage profile [3]. Data collection is also made easier through technology such as mobile apps, social media sites and search engines that directly collect data, and sensors such smart meters and room occupancy sensors that indirectly measure human activity. These technology platforms also provide a convenient framework for “human sensors” to record and broadcast data for behavioral studies, as a form of crowd sourced citizen science. This has the added advantage of engaging the broader public in STEM activities and help influence public policy.

• Research Organization: City of Los Angeles Department
• Sponsoring Organization: USDOE Office of Electricity (OE)
• DOE Contract Number: OE0000192
• OSTI ID: 1332547
• Report Number(s): DOE-USC-00192-100
• Resource Relation: Conference: Cloud Futures Workshop University of California, Berkeley, CA, USA May 7-8, 2012
• Country of Publication: United States
• Language: English