NCRTAIDS-2026-Proceeding

Blockchain-Based Tokenization Framework for Secure and Transparent Agricultural Trade

Authors: Dr. Sumathy Kingslin, Ms. K. Vaishnavi

Abstract: Agricultural markets often suffer from delayed payments, lack of transparency, and dependency on intermediaries, which significantly reduces farmers’ profit margins. This paper presents a blockchain-based tokenization framework that converts agricultural produce into digital tokens using smart contracts deployed on the Ethereum blockchain. Each token represents a verified quantity of produce and enables direct peer-to-peer transactions between farmers and buyers. The system was implemented using Solidity smart contracts deployed through Remix IDE and tested on a local Ethereum environment using Ganache. Experimental results demonstrate improved transparency, faster transaction settlement, and secure ownership transfer compared to traditional agricultural trading systems.

DOI: https://doi.org/10.5281/zenodo.18696772

Deep Learning Based Intelligent Framework for Early Detection of CF Using CNN

Authors: Mr. S. Kumaravel

Abstract: Cystic Fibrosis (CF) is a genetic disorder that severely affects the lungs and digestive organs. Timely identification of the disease can significantly enhance treatment effectiveness and patient survival. This work introduces a deep learning-based framework that utilizes Convolutional Neural Networks (CNN) to identify early signs of Cystic Fibrosis from medical images and clinical indicators. The proposed model automatically extracts discriminative features and performs classification between CF and non-CF cases. Experimental findings demonstrate strong performance and reliability, highlighting the usefulness of artificial intelligence in supporting early clinical diagnosis and healthcare decision systems.

DOI: https://doi.org/10.5281/zenodo.18696979

Multi-Task Deep Learning for Simultaneous Diabetic Retinopathy Grading and Lesion Segmentation

Authors: S Jaganathan, Dr. S. Prasath, S Jaganathan

Abstract: Diabetic retinopathy (DR) is a major microvascular complication of diabetes and a leading cause of preventable blindness worldwide, necessitating accurate and early automated diagnostic solutions. Recent deep learning–based approaches have shown promising results in DR detection; however, most existing methods focus solely on disease grading and lack lesion-level interpretability, limiting robustness and clinical reliability. To address these challenges, this study proposes a multi-task learning framework termed MTL-DRNet for simultaneous diabetic retinopathy grading and lesion segmentation from retinal fundus images. The proposed architecture employs a shared convolutional backbone to learn common representations, followed by task-specific branches for DR severity classification and pixel-wise lesion segmentation. A unified multi-task loss function jointly optimizes both objectives, enabling coordinated learning of global and lesion-level features. Experimental evaluation conducted on a publicly available diabetic retinopathy dataset demonstrates that MTL-DRNet significantly outperforms single-task, lesion-based, ensemble, and attention- driven models across standard performance metrics. The proposed model achieved an accuracy of 96.2% and an AUC of 0.98, highlighting its robustness and diagnostic effectiveness. Overall, MTL-DRNet offers an interpretable, accurate, and clinically meaningful solution for automated diabetic retinopathy screening and decision support.

DOI: https://doi.org/10.5281/zenodo.18697504

Artificial Intelligence and Big Data Analytics in Healthcare Systems

Authors: Dhineshkumar P, Harini V

Abstract: The exponential increase in digital health records, medical imaging data, wearable device outputs, and genomic information has significantly contributed to the rise of big data in healthcare. Big Data Analytics (BDA) offers advanced tools and methodologies to efficiently process, manage, and extract valuable insights from these vast and complex datasets. This paper examines how big data analytics enhances healthcare systems by improving diagnostic accuracy, enabling personalized treatment strategies, supporting predictive modeling, and optimizing hospital operations. Through the integration of machine learning algorithms and data mining techniques, healthcare institutions can uncover hidden patterns, forecast disease outbreaks, and facilitate evidence-based clinical decisions. Moreover, big data supports real-time patient monitoring, which helps reduce medical errors and improve overall health outcomes. Despite its transformative potential, challenges such as data privacy concerns, interoperability issues, and the shortage of skilled data professionals continue to hinder its widespread implementation. Therefore, this study highlights the necessity of adopting secure, scalable, and ethical data analytics frameworks to advance healthcare toward a more proactive, data-driven, and patient-centered model.

DOI: https://doi.org/10.5281/zenodo.18697789

Impact Of Artificial Intelligence On Teaching And Learning

Authors: Ms. S.Rubarani, Ms.M.Pradeepa

Abstract: Artificial Intelligence (AI) is transforming the landscape of education by reshaping teaching methodologies, learning experiences, and institutional management. From intelligent tutoring systems to automated assessment tools, AI enhances personalized learning, improves administrative efficiency, and supports data-driven decision-making. Technologies such as adaptive learning platforms, chatbots, and predictive analytics enable educators to address diverse learner needs while fostering engagement and academic achievement. However, the integration of AI in teaching and learning also raises critical concerns related to data privacy, algorithmic bias, digital equity, and the evolving role of teachers. This paper explores the multifaceted impact of AI on education, highlighting its benefits, challenges, and future implications. It argues that while AI has the potential to significantly enhance educational outcomes, its implementation must be guided by ethical frameworks, inclusive policies, and continuous professional development for educators to ensure responsible and equitable use.

DOI: http://doi.org/10.5281/zenodo.18697965

AI In Autonomous Vehicles: Sensor Fusion And Decision-Making Models

Authors: Mrs.V. Priyanka, Dr.S. Sangeetha

Abstract: Autonomous vehicles (AVs) are transforming modern transportation systems through the integration of artificial intelligence (AI), machine learning, and advanced sensing technologies. These vehicles must continuously perceive dynamic environments, interpret complex traffic conditions, and make safe driving decisions in real time. However, relying on a single sensor is insufficient due to environmental uncertainties such as noise, occlusion, lighting variations, and adverse weather. To address these challenges, autonomous vehicles implement multi- sensor fusion techniques combined with intelligent decision- making models. This paper presents a detailed study of autonomous vehicle architecture, sensor technologies, sensor fusion strategies, AI-based decision-making models, real-world applications, challenges, and future research directions. The objective is to provide a comprehensive understanding of how AI and data science enable reliable and safe autonomous driving systems.

DOI: https://doi.org/10.5281/zenodo.18698079

A Novel Surveillance System For Passenger Safety And Monitoring The Driver Behavior Through Vehicular Ad Hoc Network

Authors: Dr.M. Jaganathan, Ms.K. Sneka

Abstract: The system comprises of two Electronic Control Units (ECU) – Active Vehicle Control (AVC) to measure the driver’s driving behavior and Passenger Safety Control (PSC) for ensuring the safety during travelling in private vehicles. The AVC module is a Global System for Mobile (GSM) based ECU which consists of multiple sensor that measures the driver’s abnormal or race driving behavior and correspondingly report to the call taxi company through GSM. The PSC module enables the passenger to stop the vehicle by pressing the emergency stop button. The Vehicular Ad-Hoc Network (VANET) also known as Vehicle to Vehicle (V2V) communication shares the real time traffic related information between the vehicles to reach the destination on time thereby avoiding accidents and traffic congestion. The proposed system is found to be superior to ensure security for passengers than traditional approaches.

DOI: https://doi.org/10.5281/zenodo.18698431

Attention Mechanisms In Artificial Intelligence

Authors: Mrs. M.Poongodi, Ms. M.Janani

Abstract: Transformers and Large Language Models (LLMs) have become foundational architectures in modern artificial intelligence, particularly in natural language processing and generative modeling. Their effectiveness is deeply rooted in mathematical principles drawn from linear algebra, probability theory, optimization, and information theory. This abstract presents a mathematical perspective on the core components of transformer-based models, including vector embeddings, positional encoding, self-attention, and multi-head attention mechanisms. The probabilistic formulation of language modeling, softmax-based output distributions, and cross-entropy loss functions are examined to explain learning and inference processes. Additionally, optimization techniques such as gradient-based methods and adaptive optimizers are highlighted for efficient training of large-scale models. By emphasizing the mathematical structures that govern representation, learning, and generalization, this work provides a rigorous foundation for understanding how transformers and LLMs achieve scalability, robustness, and high predictive performance. The abstract aims to support students, researchers, and educators in developing a deeper theoretical understanding of contemporary language models.

DOI: https://doi.org/10.5281/zenodo.18698659

An Analytical Review of Deep Learning Approaches in Image Processing

Authors: Dr.P.Suresh Babu, Dr.S.Sangeetha

Abstract: Deep learning (DL) has revolutionized image processing by enabling performance far beyond traditional techniques. This survey reviews the evolution of DL-based image processing methods, from early architectures to state-of-the-art models and learning paradigms. It highlights key advancements that enhance efficiency, generalization, and robustness for analyzing complex visual data across diverse applications. Commonly used evaluation metrics are discussed to emphasize rigorous performance assessment. The survey also outlines future research directions, including quantum and neuromorphic computing, federated learning for privacy-preserving training, and the integration of edge computing and explainable artificial intelligence to address scalability and interpretability challenges.

DOI: https://doi.org/10.5281/zenodo.18766561

Green Computing in Modern Technology

Authors: N. Dhanasekaran, A. Anushya, G. Boomika

Abstract: Green Computing refers to the environmentally responsible use of computers and information technology resources. With the rapid growth of computers, data centers, and digital devices, energy consumption and electronic waste have increased significantly. Green Computing focuses on reducing the negative impact of technology on the environment by promoting energy efficiency, minimizing waste, and using eco-friendly practices throughout the life cycle of computing devices.In recent years, the demand for computers, servers, and cloud-based services has grown rapidly. This growth has resulted in higher electricity usage and increased carbon emissions. Green Computing aims to address these problems by encouraging the use of energy-efficient hardware, optimized software, and responsible disposal of electronic waste. Techniques such as power management, virtualization, cloud computing, and the use of renewable energy sources play an important role in achieving sustainable computing.Green Computing also emphasizes reducing e-waste by recycling old electronic devices and extending the lifespan of computers through proper maintenance and upgrades. Paperless communication, such as emails and digital documents, further supports environmental conservation by reducing paper consumption and deforestation. Organizations and individuals can contribute to Green Computing by adopting simple practices like shutting down systems when not in use, using energy-star certified devices, and reducing unnecessary printing.The importance of Green Computing is increasing as environmental issues such as global warming and climate change become major global concerns. By adopting Green Computing practices, industries can reduce operational costs, save energy, and protect natural resources. At the same time, it helps in creating awareness about sustainable development and responsible use of technology.This project highlights the concept, importance, and benefits of Green Computing. It shows how technology and environmental protection can work together to create a sustainable future. Green Computing is not only a technological solution but also a social responsibility that ensures the efficient use of resources while protecting the environment for future generations.

DOI: https://doi.org/10.5281/zenodo.18766889

AI and the Evolution of Modern Technology

Authors: Dr.N. Sudha, Mr. G.Senthilkumar

Abstract: Artificial intelligence (AI) is reshaping humanity’s future, and this manuscript provides a comprehensive exploration of its implications, applications, challenges, and opportunities. The revolutionary potential of AI is investigated across numerous sectors, with a focus on addressing global concerns. The influence of AI on areas such as healthcare, transpor- tation, banking, and education is revealed through historical insights and conversations on different AI systems. Ethical considerations and the significance of responsible AI development are addressed. Furthermore, this study investigates AI’s involvement in addressing global issues such as climate change, public health, and social justice. This paper serves as a resource for policymakers, researchers, and practitioners understanding the complex link between AI and humans.

DOI: https://doi.org/10.5281/zenodo.18767044

Energy-Efficient and Secure Data Transmission in Wireless Sensor Networks Using Trust-Aware Hybrid Optimization and Learning

Authors: Dr.P.Suresh Babu, Ms.K.Hemapriya

Abstract: Wireless Sensor Networks (WSNs) are widely deployed in monitoring and control applications where reliable data delivery and prolonged network lifetime are critical. However, the limited energy capacity of sensor nodes and the presence of insecure or unreliable routing paths significantly degrade network performance. Existing solutions often address energy efficiency and security independently, resulting in sub-optimal operation under dynamic network conditions. In this paper, a trust-aware and energy-efficient data transmission framework is proposed for WSNs by integrating hybrid meta heuristic optimization with a lightweight learning-based trust evaluation mechanism. The proposed approach jointly optimizes cluster head selection and routing path formation by considering residual energy, node trustworthiness, inter-node distance, and link quality. A multi-objective fitness function guides the optimization process to balance energy consumption and secure communication. Experimental evaluation using a benchmark WSN data set demonstrates that the proposed framework significantly reduces energy consumption, extends network lifetime, and improves data confidentiality and data integrity compared with state-of-the-art routing techniques. The results confirm the suitability of the proposed method for secure and sustainable WSN deployments.

DOI: https://doi.org/10.5281/zenodo.18767273

Role of Ai in Modern Education System

Authors: P. Sunitha Nandhini, P. Priyadharshini, R. Sujitha

Abstract: Artificial Intelligence (AI) has emerged as one of the most powerful and transformative technologies of the 21st century, influencing almost every sector of human life, including healthcare, industry, communication, transportation, and education. Among these, the education sector is undergoing a significant transformation due to the integration of AI-based systems and digital technologies. Traditional education systems largely follow a standardized and uniform approach to teaching and learning, where all students are expected to learn the same content at the same pace using the same methods. However, learners differ in their abilities, interests, learning styles, background knowledge, and learning speed. This mismatch between traditional teaching methods and individual learning needs often leads to learning gaps, reduced motivation, poor academic performance, and lack of student engagement. In this context, Artificial Intelligence offers a powerful solution through the concept of personalized learning.AI-based personalized learning focuses on understanding each learner as a unique individual and designing learning experiences that match their specific needs. By using data-driven algorithms, machine learning models, and intelligent systems, AI can analyze large volumes of student data such as learning patterns, academic performance, behavioral responses, engagement levels, strengths, weaknesses, and preferences. Based on this analysis, AI systems can create customized learning paths for students, ensuring that each learner receives the right content, at the right level, and at the right time. This approach shifts education from a teacher-centered model to a learner-centered model, where students become active participants in their own learning process.Artificial Intelligence in education enables adaptive learning environments in which the difficulty level, learning speed, content format, and instructional strategies are automatically adjusted according to the learner’s progress. For example, students who learn faster can be provided with advanced materials and challenging tasks, while students who struggle with certain topics can receive additional explanations, practice exercises, and supportive learning resources. This ensures that no student is left behind and no learner is held back due to a rigid system. AI-powered platforms also provide instant feedback, helping students understand their mistakes and improve continuously, which strengthens learning outcomes and confidence.Another important contribu

DOI: https://doi.org/10.5281/zenodo.18768177

Smart Energy Management System Using Ai and Iot

Authors: P. Logeswari, G. V. Sowmiya, D. Keerthana

Abstract: The rapid growth of population, industrialization, and urbanization has led to a continuous increase in global energy demand, creating serious challenges related to energy availability, cost, efficiency, and environmental sustainability. Traditional energy management systems are often inefficient, manual, and reactive in nature, leading to energy wastage, poor resource utilization, and high operational costs. In this context, the integration of Artificial Intelligence (AI) and Internet of Things (IoT) technologies offers a powerful and intelligent solution for building Smart Energy Management Systems (SEMS). These advanced systems enable real-time monitoring, automated control, intelligent decision-making, and optimized energy usage across residential, commercial, and industrial environments. IoT devices continuously monitor parameters such as electricity consumption, voltage levels, equipment performance, environmental conditions, and user behavior. This data is transmitted to centralized systems or cloud platforms, where it is stored and processed for further analysis. Based on these insights, AI systems can make intelligent decisions such as load balancing, peak demand management, energy optimization, and automated energy distribution. A Smart Energy Management System using AI and IoT enables automated control of electrical appliances, lighting systems, heating and cooling systems, and industrial machinery. These systems can automatically switch devices on or off, regulate power supply, and optimize energy usage based on real-time demand and user preferences

DOI: https://doi.org/10.5281/zenodo.18769229

Cyber Security Threats, Challenges and Emerging Defense Mechanisms

Authors: Dr. R.Senthilkumar, Mr.Boopathi.V

Abstract: Cyber security continues to be a dynamic and rapidly changing domain as new technologies emerge and the threat landscape evolves. This paper aims to explore the evolving landscape of cyber security, focusing on contemporary threats, vulnerabilities, and corresponding defense mechanisms, with an emphasis on innovative techniques and algorithms. Cyber security has become a critical concern in the digital era as organizations, governments, and individuals increasingly rely on interconnected systems and cloud-based technologies. The rapid growth of the internet, mobile computing, artificial intelligence, and the Internet of Things (IoT) has expanded the attack surface, creating new opportunities for cybercriminals. Common cyber security threats include malware, ransomware, phishing, insider attacks, distributed denial-of-service (DDoS) attacks, and advanced persistent threats (APTs). These threats target sensitive data, disrupt services, and cause significant financial and reputational damage.

DOI: https://doi.org/10.5281/zenodo.18769307

DEEP LEARNING IN NATURAL LANGUAGE PROCESSING

A Multilevel Framework Linking Generative AI, Neuroethics, And Human Resource Management In Hybrid Work Ecosystems: A Covariance-Based Structural Equation Modelling Approach

Authors: Mrs R. Ramya, Dr S. Muthumari

Abstract: The accelerated diffusion of Generative Artificial Intelligence (GenAI) has fundamentally transformed human resource management (HRM) practices within hybrid work ecosystems. While existing studies predominantly emphasise technological efficiency and performance outcomes, limited empirical attention has been paid to the neuroethical dimensions shaping employee cognition, trust, and behavioural alignment in AI-augmented workplaces—particularly within emerging economy contexts. Addressing this gap, the present study develops and empirically validates a multilevel theoretical framework integrating Generative AI capability, neuroethical governance, and strategic HRM outcomes in hybrid work environments. Drawing on the Resource-Based View, Social Exchange Theory, and Neuroethical Decision Theory, a theory-driven quantitative research design was adopted. Primary survey data were collected from 412 HR professionals and knowledge workers across Indian IT, consulting, and digital service organisations operating under hybrid work models. Covariance-based Structural Equation Modelling (CB-SEM) using AMOS 26 was employed to assess both the measurement and structural models. The findings reveal that Generative AI capability significantly enhances HRM effectiveness, mediated by neuroethical trust and moderated by hybrid work intensity. Neuroethical governance emerged as a critical mechanism through which AI-driven HR practices translate into sustainable employee engagement and organisational legitimacy. The study contributes to HRM and AI governance literature by integrating neuroethics into HR analytics discourse and offers actionable insights for managers and policymakers seeking ethically grounded AI adoption in hybrid work ecosystems.

Blochain Meets Ethereum: Unlockiing New Posibilities

Authors: P. Deepa, R. Kavitha

Abstract: The convergence of fundamental blockchain technology with the Ethereum network has ushered in a new era of decentralized innovation, moving beyond simple cryptocurrency transactions to a programmable, trustless ecosystem. By introducing smart contracts—self-executing, automated agreements—and the Ethereum Virtual Machine (EVM), Ethereum acts as a decentralized "world computer" that allows for the creation of decentralized applications (dApps) across numerous sectors, including finance, healthcare, and supply chain management. In recent years, blockchain technology has gained significant attention for its potential in various domains. However, the lack of interoperability between different blockchain platforms poses a significant challenge in meeting the demands of the modern world. To address this issue, our research focuses on unlocking blockchain interconnectivity through smart contract-driven cross-chain communication. We aim to contribute to the development of a model that enhances the functionality and usability of blockchain technology. To achieve interoperability, we explore various options and leverage the power of smart contracts.

Performance Analysis and Mitigation Techniques of Ant Colony Optimization Based Routing Protocol in Mobile Adhoc Network(Manet)

Authors: Mr. K. Mohanraj, Dr. S. Prasath

Abstract: Mobile Ad Hoc Networks (MANETs) are decentralized wireless networks where mobile nodes communicate without fixed infrastructure. Due to dynamic topology and limited resources, routing becomes a major challenge in MANET environments. Ant Colony Optimization (ACO), inspired by the foraging behavior of ants, provides an adaptive and distributed solution for routing problems. This paper proposes an ACO-based routing algorithm for MANETs that improves route discovery and path optimization. Artificial ants explore the network and update routing tables using pheromone trails. Simulation results show improved packet delivery ratio, reduced delay, and higher throughput compared to traditional routing protocols.

DOI: https://doi.org/10.5281/zenodo.19492875