No abstract

No abstract

No abstract

Internet of Things (IoT) technology has changed the educational landscape by allowing educators and administrators to turn data into actionable insight. Education organization begin to leverage solutions like cloud computing and radio frequency identification (RFID) across an IoT platform. Relative to this context, this paper proposes a five layer framework to facilitate automated student performance evaluation in engineering institutions based on smart computing concept. Student daily activity datasets are formed based on sensing capabilities of IoT nodes. Smart computing integrates hardware, software, and network technologies that provides systems with real‐time situation awareness and automated analysis. The engineering student performance per session is calculated by combining the results from sensory nodes based education data mining algorithms and student academic datasets. Moreover, based on student sessional performance score, decisions are taken by management authority to increase the reputation score of the engineering institution. The experiment comprises two sections. In first section, RFID based experimental setup is defined with objects interaction patterns. In second section, student performance score generated using proposed system is compared with manual system. The results depict that by introducing IoT in engineering education, more effective decisions can be taken to improve student learning experiences and over‐all growth of the institution.

Over the last few years, Internet of Things (IoT) has opened the doors to innovations that facilitate interactions among things and humans. Focusing on healthcare domain, IoT devices such as medical sensors, visual sensors, cameras, and wireless sensor network are leading this evolutionary trend. In this direction, the paper proposes a novel, IoT-aware student-centric stress monitoring framework to predict student stress index at a particular context. Bayesian Belief Network (BBN) is used to classify the stress event as normal or abnormal using physiological readings collected from medical sensors at fog layer. Abnormal temporal structural data which is time-enriched dataset sequence is analyzed for various stress-related parameters at cloud layer. To compute the student stress index, a two-stage Temporal Dynamic Bayesian Network (TDBN) model is formed. This model computes stress based on four parameters, namely, leaf node evidences, workload, context, and student health trait. After computing the stress index of the student, decisions are taken in the form of alert generation mechanism with the deliverance of time-sensitive information to caretaker or responder. Experiments are conducted both at fog and cloud layer which hold evidence for the utility and accuracy of the BBN classifier and TDBN predictive model in our proposed system. Graphical Abstract Student stress monitoring in IoT-Fog Environment.