International Journal of Research and Review in Applied Science, Humanities, and Technology
Not a member yet
92 research outputs found
Sort by
Energy-Efficient IoT Systems: Integrating Low-Power Protocols and Adaptive Algorithms for a Greener Future
Get PDFThe IoT has revolutionized various sectors by enhancing connectivity and automation, but the rapid expansion of IoT networks has led to significant challenges in energy consumption as well as sustainability. This paper presents a comprehensive framework for energy-efficient IoT systems, designed to reduce power consumption, extend device lifespan, and maintain high performance. The proposed framework integrates low-power communication protocols, adaptive power management algorithms, and energy harvesting techniques to optimize energy usage across IoT applications such as smart homes, industrial IoT, healthcare, along with smart cities. The methodology employed includes a literature review, system design, simulation modelling, prototype development, and field testing. The framework was tested in real-world environments to assess its impact on energy consumption, device longevity, and network performance. The results show that the proposed framework leads to a reduction in energy consumption by approximately 35% and improves device lifespan by 30-33%. These benefits were particularly prominent in smart homes and industrial IoT applications. Although there was some minor reduction in network throughput, the trade-off was minimal, ensuring that system performance remained high while achieving substantial energy savings. The study concludes that energy-efficient IoT systems can significantly reduce environmental impact and operational costs, making them essential for the sustainable evolution of IoT technologies. The framework provides valuable insights for developing greener IoT solutions that balance energy efficiency, system performance, scalability, and security. Further research is recommended to refine the framework, explore advanced energy harvesting methods, and optimize power management strategies for specific IoT domains
IoT and AI in Healthcare Management: A Review of Technologies, Challenges, and Future Trends
Get PDFThe history of the technologies breakthrough in the field of the Internet of Things (IoT) and Artificial Intelligence (AI) has assisted in the provision of various areas of life, and healthcare control can hardly be considered an exception to the rule. This technology has been creeping into the health systems and the scenario is currently augmenting the innovations in the health systems that will eventually benefit the patients with convenience in the operations and even minimisation of the cost incurred during the operations. The present paper will conduct a thorough research of the IoT and AI integration in terms of healthcare management; the essential technologies, applications, and challenges, as well the opportunities will be found. The potential to measure health in real time has emerged due to the new technology e.g. wearables and sensors and the rationale behind the AI algorithms has been transferred to diagnosis phase, treatment-planning and decision-aiding factors. The review finds the following way how such technologies can be applied to medical care in order to maximize the number of favourable outcomes such as early detection of a disease, personalized care, and prediction in hospital management. The paper will also be written in a format that captures all the issues that are raised whenever one uses IoT and AI in medical care like data security, data privacy and acceptance by the system and the laws. Finally, the paper gives the future trends of the smart healthcare solution implementation, and artificial intelligence and internet of things implications to health care systems transformation and supporting patients with improved outcomes around the world
Graph Neural Network Models for Fake News and Misinformation Detection
Get PDFThe spread of false information and counterfeit news on the Internet has become an urgent issue on the international level with serious consequences in the political, health, and social trust sectors. Traditional methods of detection, relying either on natural language processing (NLP) strategies or on machine learning models, do not consider multi-relational and multi-contextual scaffolds on which misinformation spreads. Recent advances in Graph Neural Networks (GNNs) offer a promising paradigm to learn such complicated relationships by modelling information ecosystems as graphs of users, posts and promotion paths. GNNs offer strong information-detecting strengths at scale through their use of structural and contextual dependencies in social networks. In this paper, we have critically revised the GNN-based misinformation and fake news detecting models. It talks about how the use of graph representations (including content graphs, social graphs, heterogeneous networks, etc.) can enhance detection accuracy when it combines textual, visual and relational information. The article gives an overview of popular GNNs, such as Graph Convolutional Networks (GCNs), Graph Attention Networks (GATs), and heterogeneous GNNs, and identifies them as applied to rumour detection, credibility assessment, and fake news detection early in its evolution. The implementation concerns such as scalability, graph-based on-the-fly construction, and the interpretability are discussed too. Its outcome is that the GNNs prove to be more useful than the old models because of the fact that it can produce those features that are network related, yet its computation is too complex, and that it can be adversarial is not an attribute of the real world. Future research directions also describe explainable GNNs, why they are necessary in combination with multimodal learning, and privacy-preserving detection systems. Overall, GNN-based solution is an important step forward in combating fake information since it provides a deeper insight into the functionality of interactions within the online ecosystem
A Comparative Analysis of Machine Learning Classifiers for Fake News Detection using NLP
Get PDFThe swift propagation of fake news through social media and other digital outlets represents a significant problem for information integrity and public trust in the media. We present a comparative study of machine learning models for detecting fake news in an automated fashion with a focus on a Logistic Regression, Decision Tree, Gradient Boosting and Random Forest classifiers. The paper details a straightforward methodology employing Natural Language Processing (NLP) approaches to preprocess and transform textual data: cleaning the text, removing stop words, and stemming before applying the Term Frequency - Inverse Document Frequency (TF-IDF) method for vectorization in machine learning models. Once trained on a balanced dataset of real and fake news articles, we then report comparative performance of these classifiers using key metrics including accuracy, precision, and recall, and show the results of distinguishing real from fake news. This paper details applied, interpretable, and scalable work to combat mis- and disinformation and fake news; and offer a foundation for future work or papers employing higher order techniques and datasets
AI-Based Predictive Maintenance Systems for Smart Manufacturing: A Review and Future Outlook
Get PDFPredictive maintenance (PdM) is one of the enabling technologies of smart manufacturing in which artificial intelligence (AI) is used to predict equipment troubles, or even better anticipate the failure of equipment, based on sensor data, machine learning algorithms, and digital twins. The traditional techniques of maintenance such as the reactive maintenance technique and the preventive maintenance technique will most probably introduce undue downtimes, wastages or redundancy. In comparison, AI-oriented predictive maintenance is a combination of real-time data analytics, industrial internet (IIoT), and existing deep learning technology will allow equipment to operate 24/7, minimize the risk of operations, and make decisions. It is a review article on the use of AI-based predictive maintenance in smart manufacturing. It talks about the establishment of PdM, how AI can develop real-time faults and use machine learning, deep learning and deep reinforcement learning procedures. Providing the advantages of PdM systems, the list below states with comment of automatic saving in case of downtime, of low cost and sustainability of aerospace case, case of automotive and case of energy. The issue of the heterogeneity of the data, non-standardization, threat to cybersecurity, and explainable AI is solved. Predictive maintenance and its combination with Industry 4.0, such as digital twins, edge computing, and blockchain, is another item on the list of the ways to make manufacturing systems more autonomous and resilient. The paper ends with a definition of future directions that include hybrid AI model, federated learning of collaborative PdM, and explainable AI model of trust and adoption. Smart manufacturing ecosystems can enable all this by adding intelligence to maintenance to move to more sustainable, reliable, and adaptable operations as defined in Industry 5.0
Comparative Review of Hydrological Models for Runoff Estimation: A Focus on SCS-CN, TOPMODEL, and VIC Approaches– A Review
Get PDFAccurate runoff estimation is essential for effective watershed management, flood risk mitigation, and sustainable water resource planning. Over the decades, a wide range of hydrological models have been developed, differing in complexity, data requirements, and spatial–temporal resolution. This review provides a comparative evaluation of three widely used models—the SCS-Curve Number (SCSCN) method, TOPMODEL, and the Variable Infiltration Capacity (VIC) model with emphasis on their underlying structure, hydrological processes, applicability, and performance across various hydro-climatic and land use scenarios. The SCS-CN method, although empirical in nature, remains a preferred tool for event-based runoff estimation due to its simplicity and minimal data demands. TOPMODEL, a semidistributed conceptual model, links runoff generation to terrain-driven saturation dynamics, making it well-suited for humid and sloped watersheds. On the other hand, VIC, a semi-distributed, physically-based model, enables large-scale and climate-sensitive hydrological simulations by coupling water and energy balances within a grid-based framework. This review synthesizes recent literature to outline the strengths and limitations of each model, offering guidance for researchers and water managers in selecting appropriate runoff modeling tools based on watershed characteristics, modeling objectives, and available data resources
Algebraic Structures in the Decomposition of Mixed and Multiplicative Trend-Cycle Models
Get PDFIn his study examines the algebraic foundations of mixed and multiplicative models in the decomposition of trend-cycle components within time series analysis. By leveraging algebraic structures, we explore how these models interact with seasonal patterns and variance distribution. The Buys-Ballot table is utilized to assess changes in row, column, and overall means and variances, particularly in cases where no trend is present. Our findings provide a theoretical framework for distinguishing the structural properties of mixed and multiplicative models, enhancing their application in time series modelling and forecastin
Enhancing Thermal Resilience of Epoxy/VinylesterMWCNT Nanocomposites
Get PDFThe thermal resilience of composite materials is critical for applications in extreme environments, where stability under high temperatures and oxidative conditions is paramount. This research explores improving the thermal stability and resistance to heatinduced oxidative degradation in epoxy/vinylester matrix composites by reinforcing them with multi-walled carbon nanotubes (MWCNTs). Adding MWCNTs to the polymer matrix notably enhances the nanocomposites' thermal characteristics, such as their degradation temperature and resistance to oxidation. To assess the thermal stability and degradation patterns of these composites under accelerated aging, various experimental methods, including thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC), were utilized. Results indicate that MWCNTs act as effective reinforcing agents by promoting a more stable crosslinked structure, enhancing the material’s ability to resist oxidative degradation at elevated temperatures. The study also examines the effect of different MWCNT loadings on the thermal properties, providing insight into the optimal reinforcement concentration for maximum performance. The findings demonstrate that the epoxy/vinylester-MWCNT nanocomposites offer a promising approach to improving the thermal resilience of polymeric materials for high-performance applications in industries such as aerospace, automotive, and electronics
Recent Advances in Machine Learning for Business Process Optimization: A Systematic Review
Get PDFMachine learning (ML) technology has been swiftly turning out to be the appropriate procedure of harmonizing the business activity in the cross-industrial environment. As people are getting more exposure to the big data and introducing new advancements in the possibilities of the internet, the use of ML has been solely undertaken with the aim of enabling organizations to do so in order to become more efficient and effective in their businesses by performing and making decisions. This systematic review touches on the given topic by discussing the emerging trends in the use of ML in optimization of business processes with specific mention of the importance that what it has in the operations management, supply chain management, marketing, human resource management as well as customer service. The key conclusions of the recent studies are generalized in the article and it was examined what ML-algorithms are most widespread and whether they are difficult to apply and what is beneficial in their activity. The review also predictive assumes that deep learning, reinforce learning and predictive learning would be more important in simplification of business processes as well as organisational competitiveness of the organisation. The results illustrate that ML would possess possibility to transform the likelihood of the business optimization on its way to the automation of the decision making procedure, and initiate the allocation of the resources, as well as increase the total endeavours of productivity. But the issue of privacy of the data, the lack of experts and the interface of ML systems with legacy are significant obstacles on the way to large-scale deployment. The future research directions in the field were outlined as the results of the paper in which the arguments about the necessity in the development of the extractable and understandable ML models in the business were indicated
Optimizing Supply Chain Performance Using AI and Machine Learning: A Predictive Analytics Approach
Get PDFThe paper presents the manner in which AI and ML have reshaped supply chain management (SCM) by making demand predictions, controlling inventory, reducing logistics costs, and controlling risks. It points out the opportunities of predictive analytics in enhancing the performance of supply chains in different industries. The paper analyses cases to demonstrate the efficiency benefits that AI/ML can bring as well as discuss some of the challenges, like those of quality and scalability of data and scalability and compatibility of systems. Although the process of implementing AI/ML can make the operations more efficient, it is expensive and necessitates clear planning and technological and people resources. The paper established that AI and ML have significant potential to provide businesses with a competitive advantage in the new global economy