Skip to content
Log in
Sponsored AI Networking Cookbook: Practical recipes for AI-assisted network automation and development AI Networking Cookbook: Practical recipes for AI-assisted network automation and development

Similar papers

Conference Aug 2026

IoT-Enabled Hybrid SVM–Random Forest-Based Model for Real-Time Air Pollution Prediction in Smart Cities

In smart cities, air pollution has grown to be a serious problem that has an impact on both environmental sustainability and human health. In order to predict air pollution in real time, this study suggests an Internet of Things-enabled hybrid model that combines Random Forest (RF) and Support Vector Machine (SVM). IoT sensors are used to gather environmental data, including temperature, humidity, and particle matter. The hybrid model makes use of RF’s ability to increase prediction accuracy through ensemble learning and SVM’s ability to handle high-dimensional input. With an accuracy of 96.5% and a lower RMSE of 0.24, experimental findings show that the suggested model performs better than conventional machine learning models. The technology helps decision-makers put timely control measures into place by facilitating effective monitoring and early pollution level forecast. This method helps create intelligent and sustainable urban landscapes while improving prediction dependability.

Gayatri Hegde, DR. Arun Kumar Saripalli, A. Angelin Stefi et al. · 0 citations
Open access Sep 2026

Research on Intelligent Safety Management Innovation Enabled by Artificial Intelligence and Forklift Internet of Things: A Data-Driven Risk Prediction Perspective

To address the prominent pain points of high accident frequency, lagging hazard early warning, and extensive empirical management in traditional forklift operation safety management, this paper proposes a full-process intelligent safety management innovation framework enabled by artificial intelligence (AI) and forklift Internet of Things (IoT), from the perspective of data-driven risk prediction. Against the background of deep integration of Industry 4.0 and intelligent logistics, the framework takes industrial safety governance theory and systems engineering methodology as theoretical guidance, and realizes deep integration of multi-source forklift IoT perception data, standardized full-link data governance, and AI-driven quantitative risk prediction models through a four-layer edge-cloud collaborative architecture. On the basis of data standardization and quality control, it constructs four core mathematical models: multi-dimensional operation risk assessment, improved XGBoost collision risk prediction, dynamic early warning resource scheduling optimization, and comprehensive safety performance evaluation, and builds a closed-loop management mechanism of "perception-analysis-warning-disposal-evaluation". Experimental validation based on real large-scale warehouse logistics scenarios shows that the framework significantly improves data integrity and consistency, reduces the overall accident rate of forklift operations by more than 65%, shortens the response time of safety incidents by over 80%, and enhances the scientificity and foresight of safety management decisions. The research results provide a feasible technical implementation path for the digital and intelligent transformation of industrial site safety management, and have important reference value for promoting the application of AI and IoT technologies in the field of industrial safety.

Unknown authors · 0 citations
Aug 2026

An integrated internet of things and machine learning framework for real-time wildfire monitoring and prevention

Forest fires are among the most destructive natural hazards, posing significant threats to ecosystems, infrastructure, and human life. In Mediterranean regions such as Algeria, the frequency and intensity of wildfires have increased due to climate change and human activities. This article proposes a cloud-centric hybrid Internet of Things (IoT) and machine learning (ML) framework for intelligent forest fire monitoring and prevention. The proposed system integrates distributed IoT sensor nodes equipped with temperature and humidity sensors that continuously collect environmental data and transmit them to a central gateway through NRF24L01 communication modules, while long-range communication with the cloud platform is achieved using a SIM808 cellular module. To enhance predictive capabilities, the framework combines real-time IoT sensing data with complementary meteorological variables, including wind speed and rainfall, obtained from external meteorological services during dataset construction. Six supervised ML models—logistic regression, decision tree, random forest, XGBoost, LightGBM, and CatBoost—were evaluated using historical Algerian wildfire data (2000–2003) together with a recent dataset collected in 2024. Experimental results show that XGBoost achieved the highest overall predictive performance with a test accuracy of 98.75% and an F1-score of 98.97%, while Random Forest and CatBoost also demonstrated robust and stable performance. Logistic Regression achieved competitive results with significantly lower computational cost, making it suitable for resource-constrained IoT environments. The proposed hybrid IoT–ML framework enables early wildfire risk assessment and supports proactive decision-making for forest management. These findings demonstrate the potential of integrating IoT sensing, meteorological information, and machine learning to support sustainable environmental monitoring within ambient intelligence systems.

Ali Kourtiche, Souad Belhia, Mahmoud Fahsi et al. · 0 citations
Review Open access Sep 2026

Comparative Review of Intelligent Flood Monitoring and Risk Assessment Systems: An IoT–Random Forest Perspective

Flooding remains one of the most destructive natural hazards worldwide, causing significant loss of life, damage to infrastructure, and socioeconomic disruption. Recent advances in the Internet of Things (IoT), wireless communication, and machine learning have enabled the development of intelligent flood monitoring systems capable of supporting real-time environmental monitoring and timely decision-making. This study presents a comparative review of recent intelligent flood monitoring systems with the objective of evaluating current technological approaches and identifying effective design strategies for flood risk assessment. The review systematically compares existing studies based on key criteria, including environmental sensing parameters, IoT architectures, wireless communication technologies, machine learning techniques, flood prediction capabilities, web-based monitoring platforms, and automated early warning mechanisms. Particular attention is given to an integrated framework that combines IoT-based environmental sensing with the Random Forest machine learning algorithm and its relative strengths when compared with other reported approaches. The comparative evaluation indicates that integrating multiple environmental sensors with Random Forest models generally provide improved predictive reliability, greater robustness to heterogeneous environmental data, reduced susceptibility to overfitting, and enhanced support for real-time flood risk classification. Furthermore, the incorporation of web-based monitoring interfaces and automated alert mechanisms contributes to more efficient dissemination of flood information and improved emergency preparedness. The findings provide a comprehensive overview of recent technological developments and highlight the value of integrating IoT sensing and Random Forest-based analytics as a practical direction for developing reliable, scalable, and intelligent flood monitoring and risk assessment systems. 

Unknown authors · 0 citations
Open access 2026

APPLICATION OF IOT AND AI FOR REAL-TIME FLOOD EARLY-WARNING AND RESILIENCE MANAGEMENT IN IBADAN METROPOLITAN AREA

Urban flooding poses a persistent threat to lives, infrastructure, and economic activities in rapidly growing African cities, yet early-warning systems in many contexts remain reactive and data-poor. This study develops and empirically evaluates an integrated Internet of Things (IoT) and Artificial Intelligence (AI)–based flood early-warning prototype for the Ibadan metropolitan area, Southwest Nigeria. Real-time hydrological data were collected through strategically deployed rainfall and water-level sensors across flood-prone communities and integrated with historical datasets to train an artificial neural network model for flood prediction. Geospatial analysis was used to map flood-risk hotspots, while system performance was assessed using accuracy, precision, recall, and warning lead-time metrics. Results show that the integrated IoT–AI system achieved prediction accuracy exceeding 90% and increased average warning lead time by nearly threefold compared to conventional monitoring approaches. Stakeholder validation further confirmed high usability and operational relevance for disaster response agencies and local communities. The findings demonstrate that locally developed, low-cost smart technologies can substantially enhance urban flood preparedness and resilience. The study provides a scalable model for climate adaptation planning and supports evidence-based integration of smart early-warning systems into urban disaster risk management frameworks in Nigeria and similar developing-country contexts.

N. Adeleye, B.A. Akindele · 0 citations
Open access Aug 2026

AIOT-Based Predictive Safety Framework for Underground Coal Mining: Integrating PINNs, Wearable Sensors, and Digital Twins in Indian Contexts

Underground coal mining remains one of the most hazardous industrial activities worldwide, particularly in emerging economies where complex geological conditions, methane emissions, roof instability, dust exposure, and equipment-related accidents continue to threaten worker safety. India, the world's second-largest coal producer, operates numerous underground mines under challenging geotechnical and environmental conditions. Despite significant advancements in mechanization and monitoring technologies, accident investigations indicate that a substantial proportion of mining incidents remain attributable to delayed hazard detection, fragmented monitoring systems, and limited predictive capabilities. Conventional safety management approaches are largely reactive, relying on threshold-based alarms and post-event analysis rather than proactive risk prediction. This study proposes a Mining 5.0-oriented intelligent safety framework that integrates Artificial Intelligence of Things (AIoT), Physics-Informed Neural Networks (PINNs), wearable sensing technologies, and Digital Twin models for real-time hazard prediction and decision support in underground coal mines. The proposed framework combines data from methane sensors, temperature sensors, air velocity monitors, geotechnical instruments, equipment health monitoring systems, and wearable devices measuring worker location, physiological status, and environmental exposure. These heterogeneous data streams are fused within a Digital Twin environment that continuously replicates underground mine conditions. PINNs are employed to incorporate ventilation physics, methane transport dynamics, and geomechanical principles into machine-learning models, thereby improving prediction accuracy and interpretability under sparse or uncertain data conditions. The study identifies critical research gaps in existing mine safety systems, including inadequate integration of physical laws with AI models, limited utilization of worker-centric sensing technologies, and the absence of comprehensive Digital Twin platforms for proactive safety management. To address these gaps, a socio-technical framework is developed that enables continuous risk assessment, predictive analytics, and human-AI collaborative decision-making. The proposed approach is expected to enhance situational awareness, reduce accident probability, improve emergency preparedness, and support sustainable Mining 5.0 transformation in India.

S. Kushwaha, R. Chaurasia · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.