The growing availability of digital behavioral data in social computing systems generates both opportunities and challenges for the data-driven study of collective behavior in social systems. Using examples from real social systems, this opening talk will highlight some of the open challenges in data science and network science and I will present first steps of how we can address them. I will further discuss how we can apply data science and network analysis techniques to latrge corpora of found data to test theories from the social sciences.

Social Cyber-security is an emerging scientific area focused on the science to characterize, understand, and forecast cyber-mediated changes in human behavior, social, cultural and political outcomes, and to build the cyber-infrastructure needed for society to persist in its essential character in a cyber-mediated information environment under changing conditions, actual or imminent social cyber-threats. Social cyber-security is an inherently computational social science in which news methods and theories predicated on a deep understanding of social behavior and the methodology of high dimensional network analysis, simulation, and machine learning. The methods and theories being developed for social cyber-security: a) take the socio-political context into account methodologically and empirically; b) are predicated on issues of influence, persuasion, manipulation, and theories that link human behavior to behavior in the cyber-mediated environment; and c) are focused on operational utility rather than just improving scores for machine learning algorithms or theory testing. In this presentation the basis of social cyber-security is presented, then key developments related to information operations in social media are used to illustrate the research challenges and operational goals in this area. Examples relate to the spread of disinformation, the use of bots in elections, and discrediting campaigns are presented. The need to move beyond detection and classification to dynamic operation is discussed, as is the limit of current classification systems in this area. Examples are presented that show how change in social and knowledge networks both capture the spread and impact of disinformation and the activities to counter it.

(Anti)Social Dynamics in Online Communities

TBA

Modeling Minorities in Social Networks

Social organizations should be, at the same time, robust to withstand shocks and adaptive to cope with change. To gain a fundamental understanding of these seemingly contradictory requirements, we utilize data science, network science and social science. Data driven models of network ensembles allow us to quantify potentially attainable states of social organizations, and to measure significant deviations. Our approach leads to a dynamic reformulation of stability in contrast to established concepts of equilibria. Applying controllability theory to social networks, we model socially desirable system states by influencing nodes and their interactions. Our methods of quantifying social dynamics open new perspectives for, and invite to rediscuss, computational social sciences.