Automatic Generation of Assessment Objects and Remedial Works for MOOCs

In the MOOC environments, the students feel to be alone in the process of choosing courses leading to their learning needs and work objectives. They perceive also to be controllers of their progresses with respect to calendars, fruition, assessment results. Students come into the MOOC environments to develop or enhance professional competences, to earn formative credits and to achieve certifications to get more employment opportunities, but the statistics underline high level of drop-out and few released useful credits and final certifications. These problems are mainly related to the difficulty to guarantee the “teaching presence” in courses with thousands of learners having different background and to the ineffective assessment methods for a meaningful learning process looking at the objectives and giving feedbacks for individual learning paths construction. The work, in particular, exploits the adaptation and personalization features of IWT platform in order to provide ARWE (Adaptive Remedial Work Environment) in order to fill the lack of a one-to-one tutoring mitigating the drop-out problem in MOOCs. The main original contribution of this work concerns the definition of an approach to automatically generate quizzes, exploiting a semantic-based method, in order to populate the e-Testing tool existing in ARWE, decreasing, de facto, the effort for instructors in the assessment authoring phase.