Computing and Informatics

The analysis and evaluation of computer-supported collaborative activities is a complex and tedious task. However, it is necessary in order to support collaborative scenarios, to scaffold the collaborative knowledge building and to evaluate the learning outcome. Various automated techniques have been proposed to minimize the workload of human evaluators and speed up the process. In this study, we propose a memory based learning model for the analysis, classification and evaluation of collaborative activities that makes use of time series techniques along with logfile analysis. We argue that the classification of collaborative sessions, with respect to their time series attributes, may be related to their qualitative aspects. Based on this rationale, we explore the use of the model under various settings. The results of the model are compared to assessments made by expert evaluators using a rating scheme. Correlation and error analyses are further conducted.