Search

Clustering of educational data allows similar students to be grouped, in either crisp or fuzzy sets, based on their similarities. Standard approaches are well suited to identifying common student behaviors; however, by design, they put much less emphasis on less common behaviors or outliers. The approach presented in this paper employs fuzzing clustering in the identification of these outlier behaviors. The algorithm is an iterative one, where clustering is applied, outliers identified, the data restricted to the outliers, and the process repeated. This approach produces a clustering that is crisp between each iteration and fuzzy within. It arose as a consequence of trying to cluster student progress trajectories in an adaptive learning platform. Included are results from applying the repeated fuzzy clustering algorithm to data from multiple courses and semesters at the University of Central Florida, (N=5,044). [For the full proceedings, see ED599096.]

Charles Sturt University makes its underpinning technical curriculum available to its students using an on-demand online system they call their Topic Tree. The tree is a directed acyclic graph where nodes represent topics to be learned, and edges represent the prerequisite relationships that exist between the topics. Branches on the topic tree represent concentrations in an area of knowledge, sub-branches (water quality, fluid mechanics, etc.) represent distinct subsets of knowledge - specialty. Delivery of the technical content is in three-hour modules, and students are free to choose the order in which they engage with these topics. Previous work has identified that students engage with the on-demand curriculum much as they engage with on-demand entertainment platforms such as Netflix, completing long sequences of topics with short periods between them - the traditional "binge" model of consumption. This paper presents a more fine-grained analysis of students' pathways through the topic tree, focusing on the distance between successive topics completed by the students. Students' progress is characterized by a three-dimensional framework - time, distance, and purpose. In general, pathways through the tree fall into one of four patterns: - Forward movement along a branch of the tree, - Movement backward along a branch of the tree, - Repeating the same topic, - Switching to a different branch of the tree (backward distance to the junction of the branches combined with a forward distance along the new branch) Different students engage with the topic tree using different combinations of these pathways, distance absolute distance traveled through the topics, and different time gaps between activities on the topics. This paper will identify the different combinations that can be found in the student log data. [ABSTRACT FROM AUTHOR]