Towards Optimization of Macrocognitive Processes: Automating Analysis of the Emergence of Leadership in Ad Hoc Teams

Abstract : An important focus for technical research related to machine learning has been to address not only generalization across sub-community structures within a hierarchical dataset, but also accommodating changes over time in a longitudinal dataset using evolving behavior models. We have two prototype models built and working and are extending that work to make it more scalable to larger datasets. We completed a highly scalable model, being able to be applied to networks with millions of users. We validated the model on 3 different data sets from Massive Open Online Courses and found that the sub-community structure identified by the algorithm was predictive of differences in dropout rate between subsets of students.