Sensemaking in visual analytics attempts to understand how a tool is being used in order to better hand off data for collab- oration and document the formation of research hypotheses. Conventional approaches take advantage of human experts to identify periods of sensemaking using theoretical models or utilize verbalized/written out rationale provided by the users. However, these approaches can be inefficient and inaccurate since they heavily rely on subjective human reports. In this research, we aim to understand how data-driven techniques can automatically identify changes in user behavior (inflection points) based on user interaction logs collected from eye tracking and mouse interactions. We relay the results of a supervised classification system using Hidden Markov Models to automatically predict changes in a visual data analysis of a cyber security scenario. Using cross validation, we explore the effects of different interaction behaviors. Preliminary results indicate a 70% accuracy in identifying inflection points. These preliminary results suggest the feasibility of data-driven approaches in furthering our understanding regarding sense- making and interaction provenance in visual analytics.