The present study analyzes modern optimization methods for HSN and thoroughly examines the 'Green hydrogen supply chain’GHSC, including transportation, production, storage and consumptio
Recent Trends in Optimizing Hydrogen Supply Networks 1
Devanshi Srivastava, Department of Chemical Engineering, Harcourt Butler Technical University, Kanpur, India
Adarsh Kumar Arya, Department of Chemical Engineering, Harcourt Butler Technical University, Kanpur, India
Toufik Mzili, Chouaib Doukkali University, Morocco
Murali Pujari, Department of Chemical Engineering, University of
Petroleum & Energy Studies, Bidholi, India
The increasing dependence on fossil fuels leads to higher greenhouse gas emissions and environmental pollution, highlighting the urgent need for a transition to renewable energy sources Hydrogen is emerging as a promising alternative to fossil fuels, but the high costs associated with the hydrogen supply network (HSN) have prompted researchers to explore optimization techniques This study focuses on modern optimization methods for the hydrogen supply chain (GHSC), which encompasses transportation, production, storage, and consumption, with a particular emphasis on metaheuristic optimization (MO) applications It identifies the challenges at each stage of the supply chain and evaluates how MO techniques can address these issues, while also exploring multi-objective methodologies for optimizing related problems.
Bio- Inspired Algorithms- Based Machine Learning and Deep Learning
Bio- Inspired Algorithms- Based Machine Learning and Deep Learning
Shugufta Fatima, Stanley College of Engineering and Technology for
C Kishor Kumar Reddy, Stanley College of Engineering and
Marlia Mohd Hanafiah, Universiti Kebangsaan Malaysia, Malaysia
Recent advancements in deep learning (DL) and machine learning (ML) have revolutionized healthcare by improving patient care, diagnosis, and treatment Bio-inspired algorithms, which emulate natural processes, are gaining traction for their potential to enhance ML and DL models in this sector This paper examines the current research directions and challenges in applying bio-inspired algorithms to healthcare ML and DL, focusing on their role in feature selection while addressing limitations such as scalability, interpretability, and robustness against noisy healthcare data Additionally, ethical considerations in sensitive healthcare contexts are discussed By fostering interdisciplinary collaboration and innovative algorithmic strategies, we aim to overcome these challenges and fully harness the capabilities of bio-inspired algorithms to transform healthcare delivery, leading to better patient outcomes, personalized treatment plans, and more accurate diagnoses.
Current Neuroinnovative Techniques With Machine Learning Algorithms in the Diagnosis and Classification of Neurodegenerative Diseases 61
Esra Demır Unal, Medical Faculty, Yenimahalle Training and Research Hospital, Ankara Yıldırım Beyazıt University, Turkey
Neurodegenerative diseases, including Alzheimer's disease (AD), Parkinson's disease (PD), and amyotrophic lateral sclerosis (ALS), are characterized by progressive neuronal dysfunction and protein accumulation, leading to significant disability The rising incidence of these conditions has prompted the development of innovative neurotechnologies aimed at early detection, classification, and treatment In recent decades, machine learning (ML), a subset of artificial intelligence, has been leveraged to create various computer-aided software programs for medical applications ML encompasses techniques such as neural networks, support vector machines, decision trees, random forests, logistic regression, k-nearest neighbors, multi-layer perceptrons, Gaussian mixture models, and boosting methods, which have been utilized to enhance the evaluation, diagnosis, and classification of AD, PD, and ALS.
Advanced Polyp Segmentation Using U- Net Architecture: A Review 91
Oumnia Sabri, Chouaib Doukkali University, Morocco
El mehdi El Aroussi, Chouaib Doukkali University, Morocco
Karim Abouelmehdi, Chouaib Doukkali University, Morocco
Early detection of polyps in colonoscopy images is vital for preventing colorectal cancer, a major cause of cancer-related deaths globally Accurate segmentation of polyps is essential for effective detection, and this paper explores the use of deep learning, particularly the U-Net architecture, for this purpose The U-Net's encoder-decoder structure allows it to efficiently gather contextual information while preserving spatial features, making it well-suited for polyp segmentation Evaluations using benchmark datasets demonstrate the effectiveness of U-Net models in accurately identifying polyps in colonoscopy images, highlighting its potential as a computer-aided tool for enhancing early cancer diagnosis.
Venkat Narayana Rao, Sreenidhi Institute of Science and Technology, India
Shiva Kashyap Yellavajhala, Sreenidhi Institute of Science and
Muddasani Harshith, Sreenidhi Institute of Science and Technology,
K Siva Kumar Gowda, Sreenidhi Institute of Science and Technology, India
Wastewater Stabilization Ponds (WSPs) are recognized for their efficiency in wastewater treatment, especially when designed with baffle walls (BWs), which enhance contaminant removal and reduce space requirements despite increasing construction material usage A comprehensive analysis using typical methodology (TM) design worksheets confirmed that BWs significantly decrease both the area of WSPs and hydraulic retention time (RT) Further optimization within the mathematical modeling (MM) framework achieved additional reductions in WSP area and RT, alongside a 5% decrease in concrete volume compared to TM, utilizing the interior-point approach in MATLAB and the generalized reduced gradient (GRG) algorithm in MS Excel Solver These results underscore the potential of MM for optimizing WSPs, while suggesting that alternative algorithms may yield even better outcomes.
Advanced Studies on Hydrogen Energy Using Computational Chemistry 167
Devanshi Srivastava, Harcourt Butler Technical University, India
Adarsh Kumar Arya, Harcourt Butler Technical University, India
The article explores recent optimization models for hydrogen production and transportation, highlighting the essential role of optimization in supporting hydrogen expansion It addresses the challenges posed by reliance on non-renewable energy sources, such as fossil fuels, which contribute to energy issues, climate change, and environmental degradation In contrast, hydrogen offers a clean and efficient energy alternative due to its abundance The paper emphasizes the necessity of systematic techniques, including optimization, to facilitate the growth of hydrogen infrastructure, while also discussing advancements in hydrogen transportation, storage, and generation.
Enhancing Wastewater Stabilization Ponds for Treating Domestic Wastewater Using Machine Learning 139
T Venkat Narayana Rao, Sreenidhi Institute of Science and Technology, India
Shiva Kashyap Yellavajhala, Sreenidhi Institute of Science and
Muddasani Harshith, Sreenidhi Institute of Science and Technology,
K Siva Kumar Gowda, Sreenidhi Institute of Science and Technology, India
Wastewater Stabilization Ponds (WSPs) are recognized for their efficiency in wastewater treatment, particularly when baffle walls (BWs) are integrated into their design BWs significantly reduce space requirements and enhance contaminant removal, though they increase the amount of construction materials needed A comprehensive analysis using typical methodology (TM) design worksheets confirmed that BWs lead to a substantial decrease in both the area of WSPs and hydraulic retention time (RT) Further optimization within the MM framework resulted in additional reductions in WSP area and RT, along with a 5% decrease in required concrete volume compared to TM, achieved through the interior-point approach in MATLAB and the generalized reduced gradient (GRG) algorithm in MS Excel Solver These results underscore the potential of MM for optimizing WSPs, while also suggesting the opportunity to investigate alternative algorithms for even better outcomes.
Advanced Studies on Hydrogen Energy Using Computational Chemistry 167
Devanshi Srivastava, Harcourt Butler Technical University, India
Adarsh Kumar Arya, Harcourt Butler Technical University, India
This paper explores recent optimization models for hydrogen production and transportation, highlighting their critical role in enhancing hydrogen expansion The reliance on non-renewable energy sources, such as fossil fuels, has led to significant energy challenges, climate change, and environmental degradation In contrast, hydrogen offers a clean and efficient alternative due to its abundant sources To support the growth of hydrogen, systematic methodologies must incorporate optimization techniques The study also discusses recent advancements in hydrogen transportation, storage, and generation, providing valuable insights for developing hydrogen supply chain networks.
Graph- Based Machine Learning for Enhanced Traffic Management
Deforestation and Forest Monitoring With CNN and RNN 191
Kiran Sree Pokkuluri, Shri Vishnu Engineering College for Women,
N S S S N Usha Devi, Jawaharlal Nehru Technological University,
Alex Khang, Global Research Institute of Technology and Engineering, USA
Deforestation significantly threatens global biodiversity and climate stability, making effective monitoring and management crucial To address this issue, a novel deep learning approach utilizing Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) has been proposed for forest monitoring CNNs are employed to identify deforested areas by extracting spatial features, while RNNs analyze time series satellite data to capture forest dynamics patterns This innovative mechanism enables comprehensive spatial and temporal analysis for accurate deforestation prediction.
Graph- Based Machine Learning for Enhanced Traffic Management:
S Sriram, Department of Computing Technologies, Faculty of Engineering and Technology, College of Engineering and Technology, SRM Institute of Science and Technology,
R Krishna Kumari, Department of Mathematics, Faculty of Engineering and Technology, College of Engineering and Technology, SRM
Institute of Science and Technology, Kattankulathur, India
The rapid growth of urban populations has intensified traffic congestion, prompting the need for innovative solutions This article examines the role of Graph-Based Machine Learning (GBML) in revolutionizing traffic management by enhancing efficiency, reducing travel times, and improving urban mobility for a better user experience We evaluate the effectiveness of logistic regression and random forest algorithms in classifying traffic situations, finding that while logistic regression is interpretable, random forest significantly outperforms it in terms of accuracy With its robust performance across diverse traffic scenarios, random forest proves to be the superior methodology for traffic analysis, offering enhanced accuracy, adaptability, and resilience to complex traffic patterns.
Enhanced Autonomous Driving: Various YOLO Models for Pothole
Abderrahim Waga, Moulay Ismail University, Morocco
Said Benhlima, Moulay Ismail University, Morocco
Ali Bekri, Moulay Ismail University, Morocco
Repairing potholes requires substantial financial and time investments This article explores innovative methods for pothole detection using convolutional neural networks, specifically focusing on RGB images The research evaluates the performance of three versions of the You Only Look Once (YOLO) model—YOLOv8 Nano, Small, and Medium—for effective pothole identification Additionally, the proposed system aims to alert drivers about detected potholes and suggest alternative routes, functioning in real-time across various lighting conditions, including low-light environments Key evaluation metrics include inference speed and detection accuracy, with a mean average precision (mAP@0.5) of 66%, 71%, and 51% for the YOLOv8 Nano, Small, and Medium models, respectively To enhance model training and accelerate convergence, diverse datasets were merged from multiple sources.
Unraveling Data Complexity in the Metaverse for Anomaly Detection With Python on NYC Taxi 261 ệzen ệzer, Kirklareli University, Turkey
Nadir Subasi, Kirklareli University, Turkey
In the evolving Metaverse, ensuring data integrity and security through anomaly detection is crucial This study evaluates Python-based models such as Inter Quartile Range (IQR), Median Absolute Deviation (MAD), and Local Outlier Factor (LOF) for identifying anomalies in NYC Taxi data Our analysis highlights the performance of these models in recognizing outliers within the Metaverse's complex data environment Out of 10,320 records in the NYC Taxi dataset, the IQR method identified 2 anomalies (0.019%), while the other models also contributed to the detection process.
The Synergistic Power of AIoT in Enhancing EFL Student CCT Skills in
Muthmainnah Muthmainnah, Universitas Al Asyariah Mandar,
Muliati Muliati, Universitas Bosowa, Makassar, Indonesia
Dalwinder Kaur, Manipal GlobalNXT University, Kuala Lumpur,
Misdi Misdi, Universitas Swadaya Gunung Jati, Indonesia
Ahmad Al Yakin, Universitas Al Asyariah Mandar, Indonesia
Eka Apriani, Institut Agama Islam Negeri, Curup, Indonesia
V Vasantha Kumar, Sourashtra College Autonomous, Madurai, India
This study explored the impact of integrating artificial intelligence-enhanced learning resources in English language education for undergraduate students By incorporating technologies such as GenAI, IoT, AI-ChatGPT, and platforms like YouTube into the curriculum, we aimed to create a modern and interactive learning environment This approach focuses on developing essential 21st-century skills, particularly communication and critical thinking The use of AI-ChatGPT, along with tools like mind mapping and Google IoT, not only makes learning more engaging but also significantly enhances students' critical thinking and communication abilities These insights demonstrate the transformative potential of the Internet of Things (IoT) in modern educational practices, ultimately improving student achievement.
Transforming Civic Education With ChatGPT to Boost Student Interaction and Social learning in the GenAI Era 305
Ahmad Al Yakin, Universitas Al Asyariah Mandar Sulawesi Barat,
A Ramli Rasjid Rasyid, Universitas Negeri Makassar, Indonesia
Ali Said Al- Matari, A'Sharqiyah University, Oman
Muthmainnah Muthmainnah Muthmainnah, Universitas Al Asyariah
Ahmed J Obaid, University of Kufa, Iraq
Luis M Cardoso, Polytechnic Institute of Portalegre, Portugal & Centre for Comparative Studies, University of Lisbon, Portugal
Ahmad A Elngar, Faculty of Computers and Artificial Intelligence,
To increase social interaction and encourage independent learning in Pancasila participation and group work in different classroom environments By offering empirical evidence of the efficacy of ChatGPT as a teaching tool, this research adds to the growing body of literature on AI in education We have highlighted the importance of AI in facilitating social interactions and self- directed learning, which is essential for the growth of analytical and communicative abilities Educators and lawmakers can use this research's practical insights on how to incorporate AI tools into school courses to improve student engagement and learning outcomes
To better equip their students to face the problems of the digital age, instructors should participate in ongoing professional development opportunities to learn how to use AI tools in the classroom.
Unveiling Academic Success: Harnessing Graph Machine Learning for
Polisetty Sri Hari Sai Saran, Department of Data Science and Business Systems, Faculty of Engineering and Technology, College of
Engineering and Technology, SRM Institute of Science and Technology, Kattankulathur, India R Krishna Kumari, Department of Mathematics, Faculty of Engineering and Technology, College of Engineering and Technology, SRM
Institute of Science and Technology, Kattankulathur, India
In this chapter, we delve into the utilization of graph machine learning techniques to forecast student academic performance By harnessing graph- based representations of educational data, our study endeavors to unearth underlying patterns and connections that impact student success Through a fusion of feature engineering, graph analytics, and predictive modeling, we aim to investigate the efficacy of graph- based methodologies in improving the precision and interpretability of student performance prediction systems This paper investigates the effectiveness of logistic regression, K- nearest neighbors (KNN), and a custom Graph Neural Network (GNN) model for predicting student performance in exams Our analysis reveals that the custom GNN model outperforms both logistic regression and KNN, achieving higher accuracy and efficiency in student performance prediction The custom GNN model leverages the graph- based representation of educational data, which enhances its ability to capture complex relationships and dependencies among students.
In an era defined by rapid technological advancement, machine learning and artificial intelligence (AI) have emerged as transformative forces across diverse fields From revolutionizing medicine with AI- driven diagnostics and personalized treatments to enhancing environmental science through the analysis of complex datasets, these technologies are reshaping the way we approach critical challenges Their applications in transportation—such as optimizing logistics and enabling autonomous vehicles—and education—through tools that personalize learning experiences—highlight their versatility and far- reaching impact This book explores these intersections of innovation, presenting both theoretical foundations and prac- tical implementations.
The chapters in this volume provide a comprehensive examination of how optimi- zation, machine learning, and AI are driving smarter decision- making and fostering efficiency across disciplines With a focus on real- world applications, we aim to bridge the gap between theory and practice, showcasing how these technologies contribute to solving pressing global issues From climate predictions enabled by machine learning to the integration of AI in enhancing civic education, this book underscores the potential of these tools to create meaningful change.
As editors, our goal has been to curate a collection that not only informs but also inspires We believe the case studies and research presented here will serve as a valuable resource for a diverse audience, including computer scientists, researchers, educators, and policymakers By illuminating the transformative possibilities of AI and machine learning, we hope to stimulate further exploration and innovation in these critical areas.
This book is structured to provide a nuanced exploration of the applications of machine learning and AI across four primary domains: medicine, environmental science, transportation, and education Each chapter delves into cutting- edge research and practical implementations:
1 Hydrogen Supply Networks: This chapter examines optimization techniques for designing green hydrogen supply chains It highlights key challenges in transportation and storage, focusing on strategies to enhance efficiency and sustainability in renewable energy systems.
Healthcare 6.0 focuses on the integration of bio-inspired algorithms into machine learning models to enhance patient care This chapter highlights case studies that demonstrate the effectiveness of these algorithms in facilitating early diagnosis and developing personalized treatment plans.
This section explores the application of machine learning algorithms for the early diagnosis of neurodegenerative diseases, including Alzheimer’s and Parkinson’s It highlights various classification methods and discusses their potential impact on improving healthcare outcomes.
This chapter explores advancements in medical imaging, focusing on the application of U-Net architecture for detecting colorectal cancer It highlights how machine learning significantly improves diagnostic accuracy in this field.
5 Wastewater Treatment: Focusing on industrial processes, this chapter explores how AI models are optimizing wastewater treatment operations, reducing costs, and improving environmental outcomes.
6 Stabilization Ponds: This chapter investigates innovative machine learning approaches to design and improve the performance of wastewater stabilization ponds, offering solutions for sustainable water management.
7 Hydrogen Energy: It highlights computational chemistry advancements that drive progress in hydrogen production and storage, discussing their implications for the future of clean energy technologies.
8 Deforestation Monitoring: With the application of CNN and RNN models, this chapter addresses global deforestation challenges, emphasizing the importance of machine learning in environmental conservation efforts.
9 Traffic Management: This chapter applies graph- based machine learning techniques to optimize urban traffic systems, offering insights into reducing congestion and improving city infrastructure.
10 Pothole Detection: By leveraging YOLO models, this chapter discusses real- time
11 Anomaly Detection in the Metaverse: This chapter delves into Python- based models for ensuring data integrity and reliability within virtual environments, highlighting their importance in the expanding metaverse.
12 AIoT in Education: Exploring the integration of AI with IoT, this chapter examines applications aimed at fostering critical thinking and communication skills among students in diverse educational settings.
13 AI in Civic Education: This chapter demonstrates the use of ChatGPT to enhance civic education, showcasing how AI can promote interaction and engagement in learning processes.
14 Student Performance Prediction: It investigates the potential of graph machine learning to predict and improve educational outcomes, providing insights into tailored learning strategies.
This book highlights the significant influence of machine learning and AI in tackling global challenges, showcasing various applications such as enhancing renewable energy systems and transforming education It emphasizes the potential of these technologies to revolutionize industries and enhance quality of life The research included not only illustrates current advancements but also suggests future opportunities, promoting interdisciplinary collaboration and innovation.
In conclusion, we extend our heartfelt thanks to the contributors for their expertise and dedication, which have been instrumental in bringing this book to fruition Their efforts showcase the dynamic and evolving nature of these fields We also appreciate our readers, whose engagement with this material is sure to inspire new ideas and applications.