Using learning analytics to provide personalized recommendations for finding peers

TL;DR

This paper proposes a learning analytics-based method to personalize peer recommendations in collaborative learning, aiming to enhance student support by modeling cognitive states and applying pedagogical theories.

Contribution

It introduces a novel approach that combines learning analytics with pedagogical principles to provide adaptive peer matching and scaffolding in blended learning environments.

Findings

Modeling students' cognitive states using learning analytics

Assessment of students' proximity to the Zone of Proximal Development

Framework for personalized peer recommendation and scaffolding

Abstract

This work aims to propose a method to support students in finding appropriate peers in collaborative and blended learning settings. The main goal of this research is to bridge the gap between pedagogical theory and data driven practice to provide personalized and adaptive guidance to students who engage in computer supported learning activities. The research hypothesis is that we can use Learning Analytics to model students' cognitive state and to assess whether the student is in the Zone of Proximal Development. Based on this assessment, we can plan how to provide scaffolding based on the principles of Contingent Tutoring and how to form study groups based on the principles of the Zone of Proximal Development.