Keywords: Affective State Transition, Unsupervised Learning, Cognitive State Transition, Likelihood Metric, Fuzzy Clustering
The article suggests an unsupervised model for featuring student’s learning patterns in an open-ended learning scenario. The article proceeds by generating powerful metrics to characterize the learner’s behavior and efficacy through Coherence investigation. Then, the selected features are combined through a Gaussian Hybrid Fuzzy Clustering (GHFC) that categorizes students based on their learning patterns. The proposed system features the essential behaviors of every group and associate the behaviors with ability to develop right models to gauge the learning gains between pre- and post-test scores. Also, this article explains the deployment of behavior characterization to be developed as a adaptive framework of learning behaviors.