Experimental Data for Metacognitive Skills

The increasing student’s enrollment, the cost of education and the need to make learning resources accessible any time at anywhere are propelling educational institutions to develop a pedagogical framework to support in-class activities with online learning courses. Moreover, the increasing deployment of skilled-based courses on online learning environment occasioned by the advancement of smart and wireless technology both in formal and informal paradigm means that there is a need to develop effective ways to support online learning paradigm.The transformation system brought by advancement in smart technologies can provide a learning environment that can support online learning process. The self-regulated learning process has been identified as one of the effective ways of supporting online process. The smart learning environment can be designed and developed to support self-regulated learning process. The metacognitive skills such as goal setting, time management, helpseeking, task strategy and self-evaluation can be developed in smart learning environment to provide skills enhancement tosupport online learning process. Earlier, we developed artificial neural network (ANN)-based learning agent to support the development of smart learning environment support online learning.However, to test our approach, there is a lack of dataset to implement the applicability of our methodology. In this guiding note, we present the process of generating the dataset, the trainingand testing process how, and the trained weights of the dataset can be used for predicting students’ learning style that can trigger recommendations based on the behaviour of the student inonline learning environment. This dataset is randomly generated simulating the response of students; there are possibilities that using the dataset in similar purposes in future work could resultin different outcomes and this we consider a limitation of the dataset.