Artificial Intelligence in the Assessment Process of MOOCs using a cloud-computing ecosystem

Abstract

This research shows a flow of open, flexible, and adaptable computational processes to implement a learning assessment solution incorporated into a low-cost Massive Open Online Courses (MOOCs) platform. It considers the selection of questions made by an Artificial Intelligence (AI) engine, which receives suggestions and decisions from teachers, and which the student receives, as a virtual questionnaire in a mobile application, personalizing their learning needs in real-time. The AI is based on a forecasting engine, hosted on the remote Amazon Web Services (AWS) server, the Learning Management System (LMS) controls the assessments and the Course Management System (CMS) controls the process. This computational ecosystem is a solution that reduces the cost and the need for technical support when implementing a technology related to Machine Learning and visualization for any time and place in the LMS - CMS code. To facilitate learning portability, this ecosystem is described from three ecosystem environments, LMS-CMS (Open EDX), remote server (AWS), and an application for interfaces and server communication created in Unity3D. In these environments, ten patterns interact through various micro-services to respond to the consumption mode between the Open EDX Front End and the mobile application. Fragmentation into patterns makes this research reusable and adaptable for future online learning contexts.