Automating Expertise in Collaborative Learning Environments
Noelle LaVoie, Lynn Streeter, Karen Lochbaum, David Wroblewski, Lisa Boyce, Charles Krupnick, Joseph Psotka
Journal of Asynchronous Learning Networks Volume 14, Number 4, ISSN 1939-5256
We have developed a set of tools for improving online collaborative learning including an automated expert that monitors and moderates discussions, and additional tools to evaluate contributions, semantically search all posted comments, access a library of hundreds of digital books and provide reports to instructors. The technology behind these tools is Latent Semantic Analysis (LSA), a machine learning technology that understands the meaning of words and text in ways that agree highly with human judgments. These tools were evaluated in a series of studies with the U.S. Army War College and U.S. Air Force Academy. At the Army War College, we found that the automated monitor was as accurate at identifying discussion groups in trouble as trained human instructors, and has the potential to effectively reduce the amount of time instructors spend monitoring distance learning courses. At the Air Force Academy, the expert moderator significantly improved the quality of cadets' discussion comments in a collaborative learning environment. (Contains 2 tables and 8 figures.)
LaVoie, N., Streeter, L., Lochbaum, K., Wroblewski, D., Boyce, L., Krupnick, C. & Psotka, J. (2010). Automating Expertise in Collaborative Learning Environments. Journal of Asynchronous Learning Networks, 14(4), 97-119.