Effective classroom instruction requires monitoring student participation and interaction during class, identifying cues to simulate their attention. The ability of teachers to analyze and evaluate students’ classroom behavior is becoming a crucial criterion for quality teaching. Artificial intelligence (AI)-based behavior recognition techniques can help evaluate students’ attention and engagement during classroom sessions. With rapid digitalization, the global education system is adapting and exploring emerging technological innovations, such as AI, the Internet of Things, and big data analytics, to improve education systems. In educational institutions, modern classroom systems are supplemented with the latest technologies to make them more interactive, student centered, and customized. However, it is difficult for instructors to assess students’ interest and attention levels even with these technologies. This study harnesses modern technology to introduce an intelligent real-time vision-based classroom to monitor students’ emotions, attendance, and attention levels even when they have face masks on. We used a machine learning approach to train students’ behavior recognition models, including identifying facial expressions, to identify students’ attention/non-attention in a classroom. The attention/no-attention dataset is collected based on nine categories. The dataset is given the YOLOv5 pre-trained weights for training. For validation, the performance of various versions of the YOLOv5 model (v5m, v5n, v5l, v5s, and v5x) are compared based on different evaluation measures (precision, recall, mAP, and F1 score). Our results show that all models show promising performance with 76% average accuracy. Applying the developed model can enable instructors to visualize students’ behavior and emotional states at different levels, allowing them to appropriately manage teaching sessions by considering student-centered learning scenarios. Overall, the proposed model will enhance instructors’ performance and students at an academic level.

Precision agriculture and smart farming have received significant attention due to the advancements made in remote sensing technology to support agricultural efficiency. In large-scale agriculture, the role of unmanned aerial vehicles (UAVs) has increased in remote monitoring and collecting farm data at regular intervals. However, due to an open environment, UAVs can be hacked to malfunction and report false data. Due to limited battery life and flight times requiring frequent recharging, a compromised UAV wastes precious energy when performing unnecessary functions. Furthermore, it impacts other UAVs competing for charging times at the station, thus disrupting the entire data collection mechanism. In this paper, a fog computing-based smart farming framework is proposed that utilizes UAVs to gather data from IoT sensors deployed in farms and offloads it at fog sites deployed at the network edge. The framework adopts the concept of a charging token, where upon completing a trip, UAVs receive tokens from the fog node. These tokens can later be redeemed to charge the UAVs for their subsequent trips. An intrusion detection system is deployed at the fog nodes that utilize machine learning models to classify UAV behavior as malicious or benign. In the case of malicious classification, the fog node reduces the tokens, resulting in the UAV not being able to charge fully for the duration of the trip. Thus, such UAVs are automatically eliminated from the UAV pool. The results show a 99.7% accuracy in detecting intrusions. Moreover, due to token-based elimination, the system is able to conserve energy. The evaluation of CPU and memory usage benchmarks indicates that the system is capable of efficiently collecting smart-farm data, even in the presence of attacks.

Maintaining interconnectivity between dynamic vehicular nodes and ensuring trust in vehicular networking is still a challenging issue. It is crucial for the vehicular environments to maintain a stable interconnectivity between the vehicular nodes and further to prevent the vehicular nodes from broadcasting fake messages and simultaneously protect the vehicles against tracking attacks. An efficient Trust Management System (TMS) is proposed in this paper that will ensure trustworthiness and stable interconnectivity between vehicular entities to assure road safety and reliable communication. The proposed approach incorporates a novel timestamp mechanism and block chain concepts to ensure the steady inter-connectivity between the dynamic vehicular nodes. Further, the framework leverages block chain concepts to verify the correctness of the events stored in the Road Side Units (RSUs). It also proposes a versatile hybrid trust model that uses innovative direct and recommended trust evaluation techniques to compute trust between the vehicular entities. Moreover, the framework integrates a threading mechanism to schedule message execution in direct trust evaluation, and clustering techniques to group similar messages in indirect trust evaluation. Various experimental and comparative analyses with other related studies are carried out in a simulated environment to evaluate the performance of the proposed hybrid trust model. The findings show that the proposed trust model produces an accuracy of 92% in identifying malicious nodes.

Sepsis is a life-threatening condition resulting from a dysregulated immune response to infection, often leading to organ failure and death. Early detection is vital, as delays significantly worsen outcomes. In recent years, the integration of artificial intelligence (AI), particularly Machine Learning (ML) and Deep Learning (DL), has shown great promise in enhancing early sepsis detection by identifying digital biomarkers from large-scale clinical datasets. This systematic review analyzes and synthesizes existing ML/DL approaches applied to sepsis prediction, with an emphasis on intensive care unit (ICU) settings. A total of 80 studies were included, covering diverse data sources (e.g., MIMIC-III, eICU), feature selection methods, algorithm types, preprocessing techniques, and evaluation metrics. The models ranged from traditional techniques like logistic regression and decision trees to advanced architectures such as LSTM, transformers, and ensemble methods. A key contribution of this review is the inclusion of a forest plot summarizing reported AUC and sensitivity values from selected studies, offering a comparative visual of diagnostic performance. This helps highlight the relative effectiveness of different models and provides insights into their generalizability across clinical datasets. The review also discusses challenges related to model interpretability, ethical considerations, and the lack of external and temporal validation in many studies. It further identifies trends such as the use of real-time EHR data, patient-specific model development, and explainability tools like SHAP for clinician trust. By mapping out methodological strengths and limitations in current research, this work provides actionable recommendations for future studies and clinical deployment. The review contributes to the development of more robust, interpretable, and clinically relevant ML/DL models for early sepsis detection and improved patient care outcomes.