Towards Improvements on Domain-independent Measurements for Collaborative Assessment

Assessment on collaborative student behavior is a longstanding issue in user modeling. Nowadays thanks to the proliferation of online learning and the vast amount of data on students’ interactions this modeling issue features some alternatives. The purpose is not to depend on teachers or students assessments, which either requires management effort difficult to assume (due to some students-per-teacher ratios) or depends on individual motivations (i.e. student willingness on providing explicit feedback related to collaboration). In our research we have shown that based on frequent and regular analysis of those interactions it is feasible to obtain collaborative assessments that concurs with expert valorizations. This approach relies on data mining and machine learning techniques, which are applied to infer collaborative significant student’s characteristics such as regularity, in terms of activity and initiative, and student acknowledgment of fellow-students. The advantages of the a...

• Type: Journal
• Year: 2011
• Where: EDM
• Authors: Antonio R. Anaya, Jesus Boticario
• Keywords: Data Mining, EDM 2011, Learning Management Systems, Longstanding Issue, Student Acknowledgment