Search

Recently, there has been a substantial amount of research on smart classrooms, encompassing a number of areas, including Information and Communication Technology, Machine Learning, Sensor Networks, Cloud Computing, and Hardware. Smart classroom research has been quickly implemented to enhance education systems, resulting in higher engagement and empowerment of students, educators, and administrators. Despite decades of using emerging technology to improve teaching practices, critics often point out that methods miss adequate theoretical and technical foundations. As a result, there have been a number of conflicting reviews on different perspectives of smart classrooms. For a realistic smart classroom approach, a piecemeal implementation is insufficient. This survey contributes to the current literature by presenting a comprehensive analysis of various disciplines using a standard terminology and taxonomy. This multi-field study reveals new research possibilities and problems that must be tackled in order to integrate interdisciplinary works in a synergic manner. Our analysis shows that smart classroom is a rapidly developing research area that complements a number of emerging technologies. Moreover, this paper also describes the co-occurrence network of technological keywords using VOSviewer for an in-depth analysis.

Panic disorder (PD) is considered one of the destructive ailments, with various individuals experiencing a critical functional disorder. As the range of remission for PD is achieved only between 20% and 50% with the help of regular pharmacotherapy, modern solutions are expected to deal with this issue. By taking the advantage of Internet of Things (IoT), a novel IoT-inspired behaviour monitoring framework is proposed for the analysis of panic disorder in a particular context. A quantum probability-inspired quantification measure is calculated to determine the scale of health irregularity. In addition, Temporal Data Mining (TDM) is performed for the formulation of temporal data granules to measure Individual Health Index (IHI) by utilising the Multi-scaled Gated Recurrent Unit (M-GRU) technique of deep learning. Moreover, a two-phased alert generation approach is proposed for notifying the current health condition of an individual to the concerned caretaker or medical specialist for assistive or medical services. In the comparative analysis, the proposed framework has outperformed the state-of-the-art approaches by achieving a considerable classification accuracy of 96.89% for event determination and 94.14% accuracy for health severity determination. Similarly, a considerable improvement with respect to Specificity, Sensitivity, and F-measure has been observed for the proposed framework. [ABSTRACT FROM AUTHOR]