Intelligent IoT networks create massive, diversified data streams that need real-time and massive analytical processing. Traditional cloud-centric architectures feature latency, bandwidth congestion, and limited flexibility for batch and stream data. The AI-HyECC framework for Adaptive Batch–Stream Data Analytics (ABSDA) in intelligent IoT tackles these limits. IoT sensing, edge intelligence, AI-orchestrated adaptive middleware, and cloud cognitive analytics comprise the four-layer AI-HyECC architecture. The Neural Stream Optimizer (NSO) performs real-time edge-level inference, while the reinforcement learning-powered Adaptive Task Orchestration Engine (ATOE) dynamically distributes tasks between edge and cloud based on data type, latency, and network AI-HyECC decreases latency by 38%, resource usage by 32%, and inference speed by 40% compared to cloud-only systems in experimental validation. For many IoT applications, the framework may deliver scalable, context-aware, energy-efficient analytics. Intelligent, adaptable, and resilient AI-HyECC blends batch and stream processing for next-generation IoT ecosystems.

Physical Education (PE) teaching at colleges and universities has historically faced issues such as limited personalization, ineffective monitoring, and low teaching efficiency. Advances in the Internet of Things (IoT) and Machine Learning (ML) offer smarter classrooms by combining real-time data collection and intelligent analysis. Research offers an innovative intelligent PE classroom system that includes a revolutionary Levy-Driven Polar Bear Boltzmann Machine (Levy-PB-BM), a hybrid approach that combines improved polar bear optimization (IPBO) and Boltzmann Machine (BM) to increase PE instruction accuracy and efficiency. Data were acquired from 1,000 samples of the Intelligent PE Classroom Dataset, encompassing physiological signals, motion dynamics, and ambient factors. Z-score normalization and Fourier analysis are employed to eliminate noise and extract significant information such as heart rate variability, movement dynamics, and postural signals. The data features are then optimized using an IPBO algorithm with Levy flight behaviours to improve the accuracy and reliability of selected data features for use as predictors. The optimized data features are evaluated using a BM to create probabilistic predictions of student physical active performance, and to provide adaptive feedback to each student in a personalized manner. The experimental data revealed that the prediction ratios of the proposed model were 97.14%, indicating a high accuracy in recognizing student performance behaviours and generating appropriate instructional feedback. The pedagogical evaluations reveal notable improvements in student engagement, motor skills, and instructional effectiveness. The intelligent PE classroom framework based on the proposed data-driven, personalized, and efficient approach to adapt PE pedagogy, resulting in better educational outcomes. IoT and ML-based system personalizes PE feedback, improving student engagement and outcomes. Levy-PB-BM boosts prediction accuracy (97.14%) and efficiency (95.98%) over traditional models. Enhances motor skills, fitness progress, and teacher feedback for better PE instruction.

The Industrial Internet of Things (IIoT) has brought transformative changes to industrial sectors by enabling intelligent automation, real-time monitoring, and data-driven decision-making. However, this increased connectivity and system integration have expanded the cybersecurity threat landscape, exposing critical infrastructure to sophisticated cyberattacks. Traditional Intrusion Detection Systems (IDS), particularly those relying on static signature-based detection, have proven inadequate in addressing the dynamic, heterogeneous, and resource-constrained nature of IIoT environments. Motivated by these emerging challenges, this paper presents a comprehensive review of AI-based IDS solutions developed over the past decade, with a specific focus on IIoT contexts. In doing so, it addresses key limitations in prior literature, including an insufficient focus on industry-specific constraints, a lack of attention to explainability and adversarial robustness, and an absence of a structured, future-oriented research agenda. The primary objectives are to synthesize the evolution of IDS methodologies, provide a multidimensional taxonomy that incorporates learning paradigms, deployment layers, and resource constraints, and critically assess the suitability of available datasets and benchmarks. The review highlights significant shifts from conventional machine learning techniques toward more adaptive, lightweight, explainable, and federated learning approaches. Additionally, it proposes a structured research roadmap outlining short-, mid-, and long-term directions for enhancing IDS in IIoT, such as edge-based detection, the integration of explainable AI (XAI), and the development of resilient, self-adaptive intrusion detection architectures. This paper offers a unified, actionable reference for researchers, practitioners, and policymakers seeking to strengthen industrial cybersecurity in increasingly complex and intelligent IIoT systems.

With the increasing aging of the population, the problem of old-age care is increasingly prominent. In China, 21.3% (Five-year prediction of the trend of deep aging), this study proposes a smart home-based elderly service system based on the intelligent building system. Pass "Three-dimensional linkage" Architecture (home—community—medical vertical continuity + security/environment/lighting horizontal integration + edge—cloud collaboration), integration Smoke Innovative technologies such as identification algorithm (misald alarm rate 0.7%) and enhanced learning lighting control (energy saving 35%). In order to solve"9073"The problem of home-based old-age care in the old-age care model provides technical solutions. Solving the problem of old-age care caused by population aging is of great practical significance.

The rise in complexity of large scale IoT-enabled drone logistics networks would require intelligent computing systems that can autonomously make optimal scheduling and routing decisions. Since drones are highly important in real-time delivery and surveillance services, the planning of routes and flexible scheduling are critical to the improvement of the reliability and scalability of the operations. The current means have shortcomings in terms of lack of flexibility, routing algorithms that are fixed and lack the responsiveness to the dynamic aspects of the environment like weather, air traffic and the varying need to deliver goods. To eliminate these constraints, this paper proposes the Reinforcement Learning-Based Adaptive Route Optimization (RL-ARO) model, which incorporates Deep Reinforcement Learning into the IoT-based contextual information to optimize routes and schedule. The framework allows multi-agent drones to jointly discover the best flight paths because they constantly adapt to the current network conditions, resource limitations, and energy efficiency needs. The suggested RL-ARO algorithm is implemented in a smart healthcare delivery system for the real-time medical supply and vaccine delivery environment with minimal latency and energy usage. The results of the experiment show that the framework can dramatically increase the efficiency of the route, decrease the delivery time by more than 35%, and increase the resilience of the system in large-scale and dynamical domains. The proposed method achieves the average time gap after 150 missions, energy consumption reduction (850 wh), route optimality (> 90), adaptation response time (6 s), packet deliver ratio (96–99). The RL-ARO framework optimizes drone routes using reinforcement learning and IoT data for dynamic conditions. The system reduces delivery time by 35%, increases energy efficiency, and ensures high reliability in real-time logistics. RL-ARO provides scalable and resilient drone operations, adapting quickly to environmental and operational changes.

Incorporating IoT biometric in higher education is progressively being studied to improve campus security, modernize administrative activities, and boost student services. This research sheds light on understanding students' technology readiness and perceptions of IoT biometrics in Bangladesh's higher education institutions utilizing the Technology Readiness and Acceptance Model (TRAM). 216 data were elicited through purposive sampling and evaluated by Structural Equation Modelling (SEM) in SmartPLS 4.0. The study outcomes disclose that optimism and innovativeness as positive technology readiness (TR) significantly affect both perceived usefulness (PU) and perceived ease of use (PEOU). However, discomfort, as one of the negative TR, posits negative impact on PU. Additionally, PEOU, PU, and behavioral intentions (BI) in embracing IoT biometric technology have found significant positive relationships. A comparative analysis between users and non-users implies that users tend to be more innovative and express more significant concerns about security and comfort. On the other hand, non-users show greater optimism about new technology, underscoring their advantages and convenience. Besides, a multi-group analysis underlines a significant difference in the association between PU and BI, with non-users exhibiting a more significant influence than users. As an early implication of TRAM in this context, this research provides substantial insights into the effect of technology readiness on behavioral intentions in adopting IoT biometric technology in higher education institutions of Bangladesh. The empirical evidence recommends practitioners and education stakeholders in Bangladesh stressing the diverse roles of positive and negative technology readiness, adoption intention, and psychological distinction between users and non-users in embracing IoT biometric technology.

The transformation of food production through digital tools and technologies requires not only advanced devices and processing capabilities, but also a holistic agricultural framework that unifies multiple types of stakeholders across the agri-food chain. This paper introduces the AGRARIAN platform, a stakeholder-centric, self-organizing Cyber-Physical System designed to answer the modern and future demands of sustainable agriculture, unifying farmers, developers and policy makers, all in one dynamic and scalable ecosystem. Integrating hybrid communications, lightweight services orchestration, continuous development and cross-farm decision support, the AGRARIAN platform offers a complete solution for the next generation of farm management. For the platform validation, field trials were performed for the use case of livestock monitoring in a remote area, using commercial communication links (5 G, WiFi and satellite) for real-world system evaluation. The measurements and the results analysis confirmed the capacity of the AGRARIAN platform to support the necessary functions and services, opening the way towards modern and sustainable farming.

The rapid expansion of IoT devices has introduced significant cybersecurity risks, as attackers increasingly exploit these networks’ vulnerabilities. To counter this threat, this paper presents the Privacy-Enhanced IoT Defence System (PEIoT-DS), a novel solution that emphasises data privacy while delivering high-performance intrusion detection. PEIoT-DS use federated learning to create a comprehensive intrusion detection model without necessitating the transmission of raw data to a central server. IoT devices only contribute model updates, which are then combined to improve the global model. While allowing devices to benefit from the network’s collective insights, this decentralised learning methodology safeguards data privacy. Using a real-world IoT dataset and two popular federated learning algorithms—Federated Average and Federated Average with Momentum—the study assesses the effectiveness of PEIoT-DS. The findings show that, in comparison to Federated Average, Federated Average with Momentum produces faster convergence and better intrusion detection accuracy. Our PEIoT-DS approach offers a reliable intrusion detection system for IoT networks while maintaining privacy.