Social bots are automated systems, used to post information, hijack accounts, and automatically post to social media. These bots can influence and manipulate users of social media by spreading messages or misinformation. Though several bot detection systems and crowdsourcing efforts have been developed, no systematic approach exists to effectively detect and predict bot activities. IAI and its collaborators at the University of Arkansas are working to theorize methods to understand the behavior of social bots, investigate explanatory variables and their predictive value, and collect crowdsourced test data for bot detection. Predictive socio-computational models will be developed to detect bot activity by exploiting context, temporal, and network features of social media users. Bot Detection Algorithms will be developed to distinguish Bots, Bot Groups, and coordinated Bot Behavior by using additional machine learning techniques and cross-analytics. These algorithms will be capable of identifying top propaganda disseminators and polarizers while extracting influential coordination structures among bots. Adaptive machine learning will efficiently refine the models as bot behaviors and the social media landscape changes over time. The uncertainty of analytical results will be computed and interactive visualization will help analysts to drill down and filter results. The models and algorithms will be integrated into Scraawl®, IAI’s social media analytics tool, which will collect data and provide search, visualization, and advanced analytics capabilities, as a Service or DSaaS framework.