Our research program is focused on better understanding the social and pragmatic nature of conversation, and using this understanding to build computational systems that can improve the efficacy of conversation between people, and between people and computers, to the benefit of human learning. In order to pursue these goals, we invoke approaches from computational discourse analysis and text mining, conversational agents, and computer supported collaborative learning. Our research towards this end has birthed and substantially contributed to the growth of two thriving inter-related areas of research: namely, Automated Analysis of Collaborative Learning Processes and Dynamic Support for Collaborative Learning, where intelligent conversational agents are used to support collaborative learning in a context sensitive way. Our approach is always to start with investigating how conversation works and formalizing this understanding in models that are precise enough to be reproducible and that demonstrate explanatory power in connection with outcomes that have real world value. The next step is to adapt, extend, and apply machine learning and text mining technologies in ways that leverage that deep understanding in order to build computational models that are capable of automatically applying these constructs to naturally occurring language interactions. Finally, with the technology to automatically monitor naturalistic language communication in place, the next stage is to build interventions that lead to real world benefits. Our proven technology has demonstrated significant impact on learning in dozens of classroom studies in math, science, and engineering.