Scripting and orchestration of learning across contexts: A role for intelligent agents and data mining

This chapter describes a 12-week physics curriculum that engaged students as a knowledge community across contexts: in their classroom, home, neighborhoods, and in the smart classroom. In order to support the curriculum intervention, we developed two complementary technology environments that are built on SAIL Smart Space (S3)-an open-source technology framework. Using a designbased research methodology, we instantiated an orchestrational framework that included the use of social tagging and metadata. Additionally, we devised intelligent agents to support the enactment of our collaborative inquiry scripts. We identify three important structural dimensions for which intelligent agents can play a key role in the orchestration of such curricula: Content Agents, Activity Structure Agents, and Grouping Agents. We conclude with an evaluation of these agents in support of our curriculum designs and propose a set of design principles for the role of intelligent agents and data mining in supporting cross-context learning.