The Microgenetic Learning Analytics project aims to develop a new computational method for analyzing face-to-face, collaborative learning group discussions. The goal of microgenetic analysis is to characterize conceptual change over short time periods (minutes, hours, days). The challenge of our work is finding effective means for interpreting spoken language that is often fragmented and referential. While, microgenetic analysis is viewed by educational researchers as one of the most robust methods for understanding how human learning occurs, it is a labor intensive method. For this reason, it is conducted with case study research designs involving one or a few students. Our goal is to develop a computational method that will improve possibilities for performing microgenetic analysis over larger data sets, and to expand the scope of educational research questions that may be addressed with such data sets.