Proposal for the design and implementation of Miranda: a chatbot-type recommender for supporting self-regulated learning in online environments

Abstract

The use of virtual platforms as a new space where online learning occurs has experienced a progressive increase in recent years. These platforms, also known as learning management systems (LMS), bring many benefits, not only the intrinsic ones due to their virtual modality: the ease of access and availability, but also due to the large amount of data that they store with respect to student interactions. At present, these data have not yet been processed or exploited in their entirety and if they do so they could provide various indicators that would be oriented to understand the way in which knowledge is acquired, the behavior of students in order to further improve the experience of student learning on online platforms. Fortunately, platforms like Moodle are characterized by storing a large amount of data, for that reason several plugins are developed, which add extra functionalities to the platform and use learning analytics (LA) to monitor and describe the learning process. However, most plugins do not reach a prescription level, that is, they do not delve into specific actions to improve the learning process. Thus, this study proposes the design and implementation of a chatbot-type recommendation system, the proposed tool will help students in self-regulation of their learning, providing recommendations for time and sessions, resources and actions within the platform to obtain better results.