Authoring Inner Loops of Intelligent Tutoring Systems Using Collective Intelligence

Abstract

The inner loops on the Intelligent Tutoring Systems (ITS) are responsible for analyzing step-by-step resolutions and providing the necessary support to students successfully complete a task. Building adequate inner loops that address students’ struggles requires a considerable amount of time and several interactions with experts of the domain knowledge which increases the costs and the complexity to develop ITS. To address this problem, we proposed a novel approach to build inner loops using students’ collective intelligence (CI), which every interaction of a student to solve an exercise contributes to the process of authoring the ITS domain model. To evaluate our approach, we developed an ITS in a domain of numerical expressions that was used by 147 students. As a result, we observed that our approach helps to create an ITS with comprehensive and adequate inner loops using less time when compared to the time reported in the literature and with little intervention from experts. To the best of our knowledge, this is the first domain-independent approach that uses CI in the process of authoring inner loops of ITS.

This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001.