Volume 14 Issue 13

Abstract: Low-Power Wide-Area Networks (LPWANs) have emerged as a critical enabler for large-scale Internet of Things (IoT) deployments due to their long-range communication capabilities and low energy consumption. However, achieving optimal performance in LPWAN-based IoT systems requires careful network topology design to balance energy efficiency, coverage, scalability, and reliability. This paper investigates topology optimization strategies tailored for LPWANs, focusing on technologies such as LoRaWAN, NB-IoT, and Sigfox. We propose a multi-objective optimization framework that considers node placement, gateway density, data traffic patterns, and energy constraints. By applying both simulation and analytical modeling, we showcase notable enhancements in network longevity, delay performance, and packet delivery efficiency. Our results provide practical guidelines for deploying scalable and sustainable LPWAN topologies in diverse IoT applications, ranging from smart cities to remote environmental monitoring.

Abstract: Misinformation on social media has become a critical challenge, impacting public opinion, health, and democracy. Traditional text-based methods for misinformation detection often fall short because social media content is increasingly multimodal, containing images, videos, and text. This paper explores the use of multimodal AI models that integrate visual, textual, and contextual features to improve the accuracy of misinformation detection on social media platforms. We present an overview of recent advancements, propose a multimodal framework, and discuss experimental results, challenges, and future research directions.

Abstract This study presents an enhanced face detection approach leveraging classical Haar Cascade classifiers combined with advanced image preprocessing techniques to improve detection accuracy and robustness. The proposed method applies a sequential pipeline of histogram equalization for contrast enhancement, sharpening filters to emphasize facial features, and non-local means denoising to reduce image noise. These preprocessing steps enhance the quality of the input images, enabling more reliable detection of faces under varying lighting and noise conditions. Experimental results demonstrate that integrating image enhancement techniques prior to Haar Cascade detection significantly reduces false negatives and improves the clarity of detected regions. This approach offers a computationally efficient alternative to deep learning methods for real-time face detection applications, particularly in environments with suboptimal image quality. The system is well-suited for live video or camera feeds, functioning effectively across different lighting and background conditions with minimal computational requirements, making it ideal for deployment in CCTV systems and mobile devices.

Abstract: Data has become a crucial component in the digital age that dictates decision-making, innovation, and competitive edge in many industries. There is a long-standing practice of examining, or analyzing raw data to identify significant patterns and trends in data analytics, but in recent years, it has become more than just an adjunct or supportive role and has become a critical driver of transformation. This paper highlights the growing impacts of data analytics and how it has permeated many sectors, including, but not limited to, healthcare, finance, manufacturing, education, and government. We introduce and describe three types of analytics, current trends in predictive analytics, prescriptive analytics, and real-time analytics, along with the recent emergence of new technologies such as artificial intelligence (AI), machine learning (ML), and cloud computing. This paper also discusses some challenges of data analytics such as data use and privacy and security and ethical issues that we need to overcome to realize its full potential. Finally, by looking at the future data analytics holds for us, we indicate how data-driven strategies will continue to change industries, job roles, and global competitiveness over the future decades.

Abstract — The use of artificial intelligence (AI) to automatically generate code is transforming the way software is developed. By speeding up the coding process and minimizing mistakes that humans often make, AI-powered tools are helping developers work more efficiently and creatively. This paper takes a closer look at different AI techniques used for automatic code generation, including traditional machine learning methods, advanced deep learning models, and natural language processing (NLP) approaches that enable computers to understand and produce human language. Recent breakthroughs, especially with transformer-based models, have led to powerful tools like GitHub Copilot, which can assist programmers by suggesting code snippets in real time. We explore how these technologies work, their advantages, and the challenges they still face - such as handling complex programming tasks or understanding context deeply. Finally, this paper discusses open questions and promising directions for future research, as this exciting field continues to evolve quickly.

Abstract: The rapid-fire elaboration of software development methodologies has needed the relinquishment of robotization- driven practices to enhance software quality and delivery speed. Continuous Integration (CI) and Continuous Delivery (CD) are foundation practices in ultramodern DevOps channels, enabling brigades to integrate law constantly, descry issues beforehand, and emplace operations fleetly with minimum homemade intervention. This exploration paper investigates the perpetration and optimization of CI/ CD channels using Amazon Web Services (AWS), a leading pall computing platform that offers scalable and intertwined DevOps tools. The study presents a methodical approach to constructing a CI/ CD workflow using AWS native services similar as AWS Code Commit for source control, AWS Code Build for automated testing and compendium, AWS Code Deploy for operation deployment, and AWS Code Pipeline for unity. The proposed channel is tested using a sample pall-native web operation and crucial criteria similar as figure time, deployment frequency, and failure recovery time are estimated. The exploration highlights the benefits of integrating CI/ CD practices with pall structure, including bettered software trustability, reduced deployment crimes, and enhanced development dexterity. Likewise, the paper discusses security considerations, cost- effectiveness, and scalability aspects associated with using AWS- managed CI/ CD services. The findings give practical perceptivity for software masterminds, DevOps interpreters, and experimenters seeking to apply flexible and effective software delivery channels in pall surroundings.

Abstract: The exponential growth in urban populations has placed tremendous stress on existing traffic infrastructures. As a result, cities are experiencing increased traffic congestion, extended travel times, and increased levels of pollution. To address these challenges, this paper proposes a novel intelligent traffic management framework that integrates Artificial Intelligence (AI) and Blockchain technology for enhanced urban mobility and system security. The primary objective of this study is to develop a decentralized and automated traffic control system that can optimize signal timing and predict congestion in real-time while maintaining secure and transparent data communication across stakeholders. The methodology involves a multi-phase approach starting with data collection from CCTV cameras, IoT-based road sensors, and GPS-equipped vehicles. These diverse data streams can processed using Convolutional Neural Networks (CNNs) for vehicle detection and traffic density classification, and Long Short-Term Memory (LSTM) networks for time-series forecasting of congestion patterns. A private Ethereum blockchain can be deployed using tools like Ganache, where smart contracts developed in Solidity manage access control, log AI-generated decisions, and automate responses such as emergency vehicle prioritization. Integration of AI and blockchain components allows system nodes—such as smart traffic lights and control centers—to autonomously execute verified decisions. The proposed system will be implemented to improve traffic flow efficiency, minimize waiting times, and enable reliable vehicle prioritization across various traffic scenarios through AI-driven control mechanisms. The blockchain layer will be deployed to ensure tamper-proof and auditable records of all transactions and decision-making processes. In conclusion, the proposed AI-Blockchain integrated system enhances the efficiency, transparency, and robustness of urban traffic management, representing a scalable and secure solution for future smart cities.

Abstract: Within the modern advanced time, the danger of disasters—both normal and man-made—poses noteworthy dangers to organizational information judgment and operational progression. A novel concept called Catastrophe Recuperation as a Cloud Benefit (DRaaS) employments cloud computing to supply calamity recuperation arrangements that are reasonable and versatile. This consider analyzes the center thoughts of DRaaS, surveys the body of investigate and observational prove, and looks into the troubles and real-world employments of the innovation. The consider assesses the adequacy of DRaaS arrangements and looks at organizational appropriation patterns employing a mixed-method approach. Although it seems that DRaaS essentially reduces costs and speeds up recovery, problems with security, integration, and compliance persist. The primary focus areas of the proposals are future bearing research and the best practices for a successful DRaaS installation. The increasing dependence on digital infrastructure has heightened the need for efficient disaster recovery (DR) methods. Disaster Recovery as a Cloud Service (DRaaS) is researched here as an elastic, cost-saving substitute for conventional DR solutions. It is aimed at investigating how DRaaS can help industries reduce downtime and data loss, with a particular emphasis on small and medium-sized businesses (SMEs). The research utilizes a mixed-methods design, integrating comparative evaluation of cloud-based and on-premises DR models alongside case studies of organizations using DRaaS. Recovery time objective (RTO), recovery point objective (RPO), cost-effectiveness, and system robustness are measured through simulation and user-reported values. Outcomes show that DRaaS decreases RTO and RPO dramatically when compared to traditional systems, while providing increased flexibility and less capital investment. SMEs especially enjoy the pay-as-you-go approach and automated failover options. Concerns regarding data sovereignty and vendor lock-in are still significant challenges. Finally, DRaaS appears to be an efficient and feasible solution for disaster recovery in the contemporary era, particularly for companies requiring agility and cost savings. The publication recommends best practices for DRaaS implementation, including careful vendor scrutiny, compliance alignment, and hybrid deployment strategies.

Abstract: After being developed for cryptocurrency, now the blockchain technology is exploited for the benefit of various industries, among those the healthcare industry. Being secure and decentralized, the blockchain system is apt for securing sensitive health data. This study will explore the use of blockchain to solve prominent health problems, including data privacy, interoperability, fraud, and the control of patient information. We also explain the fundamentals of blockchain technology, elucidate healthcare use cases for blockchain, and weigh the benefits and drawbacks of implementing blockchain technology in real-world healthcare systems.

Abstract: Exchange rate fluctuations are a critical element in the economic performance of open economies, as they influence international trade, capital flows, investment planning, and policy development. For India, managing currency volatility is essential due to its increasing engagement in global markets. The Reserve Bank of India (RBI) frequently intervenes to regulate extreme movements in the exchange rate to safeguard macroeconomic stability. This study focuses on examining the USD-INR exchange rate in relation to four key macroeconomic variables: inflation rate, interest rate, unemployment rate, and GDP growth rate, considering both Indian and U.S. perspectives. To achieve this, two Multiple Linear Regression (MLR) models were constructed using annual data from 1991 to 2021, allowing for the assessment of each variable’s statistical influence and directional effect on the exchange rate. Alongside this, a time-series analysis was conducted using the ARIMA (Auto Regressive Integrated Moving Average) model to forecast monthly exchange rates, offering insights into future currency trends based on historical data patterns. The analysis revealed that macroeconomic indicators from the United States have a more substantial impact on the USD-INR exchange rate than those from India, underscoring the Indian rupee’s sensitivity to global economic conditions. The ARIMA (1,1,1) model emerged as the most suitable for forecasting purposes, providing reliable projections for the years 2021 and 2022.Overall, this research highlights the interconnected nature of global economies and emphasizes the importance of combining regression analysis with time-series forecasting to gain a comprehensive understanding of exchange rate behavior. The findings provide valuable input for policymakers, investors, and businesses engaged in international operations, as they navigate currency-related risks and develop informed strategies in a volatile global environment.

Abstract: Garbage collection (GC) is an essential aspect of memory management in Java that helps automate the process of reclaiming unused objects, thereby reducing the chances of memory leaks and improving overall system performance. Although Java provides default GC settings, these may not always be optimal for high-performance or large-scale systems. This paper explores the inner workings of Java's garbage collectors, including Serial, Parallel, Concurrent Mark-Sweep (CMS), and G1, and provides guidance on selecting the most suitable one based on specific application needs. It also covers advanced tuning techniques involving heap size configuration, garbage collection logs, and monitoring tools that assist in identifying memory bottlenecks. Additionally, it discusses best practices and common mistakes developers encounter when fine-tuning GC settings, with a focus on balancing latency and throughput. By understanding and applying these optimization strategies, Java developers can significantly enhance application responsiveness and minimize downtime.

Abstract: This research paper introduces an abstract avoidance system controlled via Bluetooth and voice commands using an Arduino board. Ultrasonic sensors detect obstacles, providing real-time data for the system. Users can remotely control the system's movements and navigate through complex environments using a mobile device connected via Bluetooth. Additionally, voice commands enhance usability and convenience. The integration of hardware components, including ultrasonic sensors, an Arduino board, and a Bluetooth module, along with algorithm development for obstacle detection and communication protocols, enables the system's functionality. This versatile system finds applications in robotics, automation, and smart environments where obstacle avoidance is crucial. By combining Bluetooth and voice control, this project offers an efficient and user-friendly solution for enhancing control and safety in various real-world scenarios.

Abstract: As AI becomes deeply entrenched in mission-critical domains like healthcare, finance, and government, the need for reliable, explainable, and ethically compliant AI systems has increased immensely. Most traditional AI systems exist as opaque "black boxes" wherein it is not possible to see how decisions are being made or to verify compliance with rules and ethical requirements. This transparency issue makes it challenging to hold AI systems accountable and develop confidence in the outcomes produced by them. This work presents a new framework that marries the advantages of blockchain technology with explainable AI to produce transparent and auditable AI systems. The essential properties of blockchain decentralization, immutability, and automation of smart contracts are utilized to have tamper-proof records of the whole AI life cycle. This entails data gathering, preprocessing, and training of models, updates, and inference events. These logs create an unalterable audit trail that allows regulators, users, and stakeholders to confirm the integrity and fairness of the AI models at any given moment. Moreover, the framework incorporates explainable AI methods to produce human-interpretable explanations of model outputs. This not only enhances transparency but also enables stakeholders to determine whether AI judgments are reasonable and unbiased. We provide a prototype implementation of this framework and compare its performance in a real-world case study in the healthcare industry. The findings show that the integrated system effectively strengthens traceability, establishes trust, and facilitates regulatory compliance without any decline in the performance of AI models. In summary, this study demonstrates how the integration of blockchain and AI closes essential gaps in transparency and accountability, paving the way for the responsible and ethical use of AI. The framework outlined provides a pragmatic way forward for companies wishing to implement AI technologies without diminishing public trust and fulfilling legal requirements.

Abstract: Healthcare is revolutionized by the unimaginable speed with which AI has evolved. The AI technologies, in particular, are at the forefront in diagnostics, personalization, and the simplification of administrative and R&D processes in the pharmaceutical industry. The investigation discusses the health care applications and advantages along with the problems and ethical issues that are traits of AI systems. Between major case studies and new strides in AI, technology is making health outcomes better while it is protecting patients' privacy, stopping biases, and reducing the need for human intervention.

Abstract: It has been widely observed and researched that the Personality of the students to a large extent influences their academic and non-academic performance. In recent years the students have gone through the phase of covid-19 pandemic. They got to experience variations in academic environment at educational institutes and at home.  They faced changes in their physical, behavioral, and mental health patterns. There were a lot of lifestyle changes caused by the pandemic. There is a remarkable difference in their learning preferences, attitudes, approaches and academic and non-academic parameters. The research paper aims to capture the gradual changes that were introduced with the outbreak of the pandemic. These changes occurred in both the teaching and learning styles. It is interesting to find out these variations and analyze their behavior in the post pandemic phase. A survey has been carried out to understand these factors.

Abstract: Mental health disorders such as depression, anxiety, and bipolar disorder significantly affect the well-being of individuals and often go undiagnosed due to reliance on subjective assessments. Voice data, being non-invasive and widely accessible, provides an excellent medium for detecting emotional and cognitive cues associated with mental health conditions. This research investigates the application of deep learning for analyzing vocal features to detect early signs of mental health disorders. Using publicly available datasets and spectrogram-based preprocessing, we evaluate Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and hybrid models. The results demonstrate the effectiveness of deep learning in identifying subtle vocal biomarkers and provide insights into real-time, scalable mental health screening tools.

Abstract: Cloud-based solutions provide flexibility in storage and processing capabilities, allowing for tailored adjustments as organizational needs evolve. Collaboration is fostered, enabling data sharing and teamwork among diverse users and teams. Accessibility becomes universal, harnessing the potential of big data analytics from any location with an internet connection.  The period of big data has brought about unknown challenges and openings for data processing and analytics. The volume, haste, and variety of data generated bear scalable and effective processing results. Real- time, contextual, and secure data is now charge critical.data streaming platforms (DSPs) play a vital part in simplifying access to real- time data and making it easy to exercise whenever and wherever it’s demanded. This paper explores the concept of big data processing in the cloud and its significance in enabling real- time data analytics. By using cloud infrastructure, we can process huge amounts of data assisting real-time data analysis. The paper delves into the benefits of cloud based big data processing, including elastic resource provisioning, cost optimization, and simplified data operation. Through this comprehensive analysis, the paper aims to exfoliate light on the eventuality of cloud computing in scalable and real- time big data analytics, empowering associations to decide precious perceptivity and make data- driven opinions.

Abstract: Customer reviews are very important in today's digital world for influencing potential customers and for establishing brand perception. A Natural Language Processing (NLP) technique called sentiment analysis makes it possible to automatically read textual opinions and identify whether they are very negative, negative, neutral, positive and very positive. This study explores the use of machine learning algorithms and a variety of natural language processing techniques for sentiment analysis of customer evaluation. Text preprocessing, vectorization, model training, and performance assessment using metrics like accuracy, precision, recall, and F1-score are all included in the study. The findings show that when it comes to understanding contextual sentiment in customer evaluations, deep learning model, particularly LSTM perform better than conventional machine learning models.

Abstract — The era where misinformation spreads rapidly across digital platforms, ability to distinguish between authentic and fabricated news has become a critical societal challenge. This project presents a machine learning-based approach to fake and real news detection using natural language processing techniques. Utilizing a labelled dataset comprising 6,335 news articles, the model analyzes both the title and content of each entry to accurately classify them as either “FAKE” or “REAL.” Pre-processing steps, including tokenization, vectorization, and noise removal, was applied to enhance text clarity. Multiple machine learning algorithms were evaluated, with performance measured through accuracy, precision, recall, and F1-score. The results underscore the efficacy of supervised learning techniques in automating the verification of news content, offering a scalable solution to combat the proliferation of misinformation in online media.

Abstract — A recent approach to creating and launching applications without managing servers is server less computing. This study examines how AWS Lambda and AWS Step Functions, two essential services, allow server less computing. While Step Functions assist in coordinating several processes into an easy workflow, AWS Lambda enables developers to execute code in response to events without worrying about infrastructure. When combined, they facilitate the development of applications that are scalable, effective, and economical. In this article, we describe the operation of these services, their advantages, and practical applications that demonstrate their value in resolving issues in a timely and dependable manner.

Abstract: This study explores the dual nature of social media’s influence, highlighting its transformative effects on both personal and professional interactions. The impact of social media on interpersonal relationships and communication is multifaceted, shaping how individuals interact in both positive and negative ways. On one hand, social media platforms allow users to maintain relationships across distances and geographical boundaries in real-time. These platforms support long-distance connections and faster global networking, often enhancing emotional well-being through online communities and support groups that provide information and access to valuable resources. The nature of online communication often leads to superficial interactions, where the number of connections overshadows the quality of relationships. The absence of nonverbal cues in digital communication can cause misunderstandings and misinterpretations. Additionally, privacy concerns are common, with personal information potentially exposed to unintended audiences or misuse. Excessive use of social media can also contribute to mental health issues such as anxiety and depression, often intensified by social comparison and cyber bullying. This study emphasizes the need for a balanced approach to social media use, recognizing both its benefits and challenges. As social media continues to evolve, understanding its impact on interpersonal relationships and communication is essential for maximizing its advantages while minimizing its drawbacks. This awareness can help individuals navigate the digital world, foster meaningful connections, and maintain a healthy balance between online and offline interactions.

Abstract — A component of artificial intelligence (AI), machine learning (ML) enables computers to learn from historical data, identify trends, and make judgments with little to no assistance from humans. Protecting computers, smartphones, servers, networks, and data from malicious attacks is the goal of cyber security. There are two ways that machine learning and cyber security can work together: by protecting machine learning systems and by leveraging machine learning to enhance cyber security. This combination has the potential to improve cyber security technologies, detect unknown and novel threats (known as zero-day attacks), and lessen the need for human intervention. Protecting critical data and systems becoming more difficult as technology advances quickly. In order to improve cyber security, this project intends to use machine learning to develop three distinct systems.

Abstract: In artificial intelligence, supervised learning has become a dominant paradigm that allows developments in a variety of fields, including natural language processing, speech recognition, and image classification. However, success generally depends upon the availability of large labeled datasets, which are frequently high-priced or impractical to obtain in many real-world situations—particularly in domains like security, bioinformatics, and healthcare. Few-shot learning techniques, which look for to allow models to generalize effectively from a limited number of training examples, were developed in response to the difficulty of learning from limited data. The current study explores the three main few-shot learning strategies—transfer learning, meta-learning, and data augmentation—as solutions for the supervised learning problems of small datasets. To improve models for new, smaller tasks, transfer learning makes use of knowledge gathered from large-scale tasks. Models can quickly adjust to new tasks with little data thanks to meta-learning, also known as "learning to learn." Small datasets are artificially expanded using data augmentation techniques to increase robustness and generalization. We look at how these approaches improve supervised models' the capacity for generalization, minimize over fitting, and reduce variance. This paper specifies the advantages, disadvantages, and uses of each approach through a thorough evaluation of previous studies and comparative analysis. In addition to the outcomes, hybrid approaches that combine these tactics perform better, particularly in fields with a lack of labeled data. In the final analysis, few-shot learning sets the way for a more efficient and equitable application of AI in situations with limited resources.

Abstract: This mixed-methods study explores how computer science educators (N=17) handle the use of generative AI tools like ChatGPT and Copilot. While 65% of participants reported spending less time on lesson planning and grading, 68% faced "validative overload", a newly identified issue where educators spend too much time checking AI outputs. Using cognitive load theory (Sweller, 2020), we examine how specific challenges, such as debugging AI-generated code, increase unnecessary cognitive load. Our findings show that 58% of educators lack training for AI integration 73% of AI-generated coding examples need major corrections. Validation tasks add 2.4 hours per week to the workload. We suggest a three-tiered framework for responsible AI use, focusing on pedagogical alignment, validation processes, and institutional support systems.

Abstract: The research paper introduces a novel method for monitoring environmental conditions in agricultural environments by integrating AI technologies with IoT infrastructure. It utilizes data from a variety of sensors including those that measure soil moisture, gas levels, and other environmental parameters to provide real-time condition tracking. The system uses an Arduino microcontroller, an ESP module for communication, and the ThingSpeak platform to gather, upload, and manage data from environmental sensors effectively. One of the system's core functionalities is its weather prediction module, developed in Python using a Convolutional Neural Network (CNN) to enable AI-driven forecasting. This module delivers valuable weather insights, supporting informed and proactive farm management. Additionally, the system includes an intuitive web interface that displays real-time sensor readings and predictive analytics, empowering farmers to optimize resource usage and respond effectively to environmental changes.

Abstract: The Smart Bike Monitoring System uses IoT since it is an advanced solution incorporating Internet of Things (IoT) technology into two-wheelers to improve safety, performance, and environmental awareness. Various sensors as well as devices are integrated within the system to collect real-time data. This data comes directly from the bike and from its surrounding environment. Actionable perceptions get generated, faults become detected, also key performance metrics get monitored while this data is processed as well as analyzed for optimizing bike usage. The system is remaining continuously connected to the internet for use. Therefore, a web interface or mobile application lets bike owners monitor their vehicle remotely. The car follows lanes is a core feature plus automatic brakes prevent accidents plus real-time accident and emergency alerts are sent plus the car detects engine smoke/emission plus headlights activate automatically in dense fog plus real-time alerts and notifications are delivered. This thorough IoT-based solution does greatly improve rider safety vehicle efficiency and environmental responsibility.

Abstract: Cyber security means protecting information, devices, computers, computer resource, communication devices and information stored there in from unauthorized access, use, modification or destruction. Cyber Security plays an important role in the field of information technology. Securing the information have become one of the biggest challenges in the present day. Cyber security is a way of protecting the computers, network, and other devices from cyber criminals. Cybersecurity is the practice of protecting people, systems and data from cyber-attacks by using various technologies, processes and policies. At the enterprise level, cyber security is key to overall risk management strategy. Cybercrime is a crime which includes computer and network to execute a crime. For example, unauthorized access or modify data or application, intellectual property theft, writing or spreading computer viruses etc. Whenever we think about the cyber security the first thing that comes to our mind is cyber-crimes which are increasing immensely day by day. Cybercrime may put a person or a nation security in danger and it is not good for financial health. Cybercrime, especially through the internet, has grown because computers are used in every field like commerce, entertainment and government. This paper Approaches to prevent, detect, and respond to cyber attacks are also discussed. In the current world that is run by technology and network connections, it is crucial to know what cyber security is and to be able to use it effectively. This thesis aims to develop a cybersecurity threat detection or attack detection system.

Abstract—AI is significantly impacting cardiovascular disease diagnosis and management, enhancing accuracy, speed, and early detection. AI algorithms can analyze ECGs, imaging data, and other clinical information to identify heart conditions, predict risks, and personalize treatment strategies. This includes detecting structural heart diseases like hypertrophic cardiomyopathy and aortic stenosis, as well as predicting long-term outcomes for heart failure patients. 

Abstract: Forecasting energy use has become a crucial component of contemporary smart grid systems, allowing stakeholders to guarantee system dependability, cost effectiveness, and energy efficiency. For the integration of intermittent renewable energy sources, load balancing, and real-time energy management, the capacity to predict power demand is essential. The use and relative effectiveness of five supervised machine learning algorithms Linear Regression, Decision Tree, Random Forest, XGBoost, and Gradient Boosting for predicting short-term building-level energy consumption are examined in this work. In order to train and evaluate models, we carried out a thorough preprocessing and feature engineering procedure using a large dataset that included operational, meteorological, and temporal variables. Each model was assessed using three key performance metrics: mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R²). Among the models tested, Gradient Boosting achieved the highest accuracy, with an MAE of 575 kWh, RMSE of 851 kWh, and an R² of 0.949, outperforming both traditional and advanced ensemble models. Our results highlight how well boosting strategies work for energy forecasting jobs and how crucial it is to choose models according to deployment restrictions and data properties. The knowledge gained from this research can help designers create responsive, scalable, and intelligent energy forecasting systems that are appropriate for smart infrastructure.

Abstract: Urban traffic is more stressed with the increase in traffic offenses like signal jumping, overspeeding, and helmet-less driving, which are compromising the safety of the roads and making transportation inefficient. Traditional methods of enforcement that rely on human monitoring and post-incident analysis are inadequate for real-time intervention. This paper introduces a future-proof smart traffic violation detection system driven by Edge AI and Internet of Things (IoT) technology to provide efficient, autonomous, and scalable traffic monitoring for contemporary urban environments. The envisioned system combines YOLOv5-based object detection, optical character recognition (OCR) for license plate extraction, and OpenCV for visual analysis. Edge devices like Raspberry Pi devices, along with IoT sensors, analyze and process video feeds at the point of origin, cutting latency and bandwidth consumption drastically. Besides detection, the system uses machine learning-based predictive analytics to predict hotspots of violations and peak hours, enabling authorities to implement preemptive safety measures. Real-time notification, automatic reporting, and integration with smart city infrastructure further increase responsiveness and public accountability. Field tests show high detection accuracy in various lighting and weather conditions, and the edge-IoT architecture provides cost savings and simplified deployment. This research helps in developing intelligent transport systems, providing secure, intelligent, and adaptable urban mobility solutions.

Abstract: Artificial Intelligence (AI) is revolutionizing healthcare by enabling the prediction of diseases through the analysis of medical data. Using machine learning algorithms, AI can detect patterns in patient history, genetic data, and lifestyle factors to predict conditions such as heart disease, diabetes, and cancer. These predictions help in early diagnosis, personalized treatments, and more efficient healthcare delivery. While challenges like data privacy and model transparency exist, AI holds significant potential to improve disease prevention, diagnosis, and patient outcomes.

Abstract: This research examines the impact of different data structures on the efficacy, precision, and efficiency of AI systems' decision-making. These data structures include arrays, linked lists, trees, graphs, and hash tables. Artificial intelligence's (AI) decision-making process is highly dependent on the data structures utilized to arrange, retrieve, and modify information. Knowing how data architectures affect AI decision-making is crucial as AI develops and applications get more complicated.

Abstract: Hospital readmissions are a significant concern for healthcare systems, resulting in increased costs and adverse patient outcomes. This study develops and evaluates a predictive model for patient readmission using Electronic Health Records (EHR) data. This study explores various machine learning techniques to predict 30-day hospital readmission rates, focusing on feature selection, model performance, and clinical interpretability. We employed machine learning algorithms, including logistic regression, decision trees, and random forests, to identify patients at high risk of readmission. Our model incorporates demographic, clinical, and healthcare utilization data from EHRs. Results show that our predictive model accurately identifies patients at high risk of readmission, with an area under the curve (AUC) of 0.85. The model also identifies key risk factors contributing to readmission, including prior hospitalizations, comorbidities, and medication adherence. Our findings suggest that predictive modelling using EHR data can inform clinical decision-making and reduce hospital readmissions. This study highlights the potential of leveraging EHR data and machine learning algorithms to improve patient outcomes and reduce healthcare costs.

Abstract: Early Indian trade routes It was critical to the economic and cultural development of the subcontinent. These intricate networks unified the cities, ports, and trade centers of the Indus valley to the Southeast Asia and have enabled the exchange of goods including spices, textile, metals, and medicinal plants. Modern Operations Research (OR) tools have been also employed in this study: The graph theory and shortest path problem will be used in order to analyze and model those historical trade routes. Historical sources, archaeological data, geographical reconstructions have been used to model ancient trade networks as weighted graphs (the nodes are trade centers, and the edges are routes with related distance, and risks e.g., terrain difficulty, political instability, banditry, and weather hazards). Applying such algorithms as Dijkstra’s, Bellman-Ford, and Floyd-Warshall the study is able to find the optimal paths, which might have been favoured by ancient traders, on various constraints, namely, travel time, cost, and risk. The multi-objective optimization model is also presented in the paper to consider efficiency and safety in order to capture the real-life decisions taken by traders in dynamic historical situations. The results point to the existence of proto-optimization behavior in ancient Indian trade and they offer a new interdisciplinary solution on how to relate historical geography and mathematical modeling. This study does not only unearth the strategic genius of the ancient Indian traders but also proves the evergreen applicability of OR in resolving practical issues.

Abstract: Ancient Indian Mathematics has made significant contributions to arithmetic, trigonometry and algebra, many of which continue to influence modern computational methods. Techniques such as Bhaskara I’s sine approximation and Vedic multiplication were designed for rapid mental calculations and have inspired the development of various modern algorithms. This paper explores the implementation and computational performance of two ancient Indian mathematical techniques—Vedic Multiplication and Bhaskara I’s Sine Approximation using Python. Their efficiencies are evaluated against modern numerical libraries like NumPy in terms of execution time, computational complexity and accuracy. The findings show that some ancient techniques are highly efficient for specific tasks even today. This study connects traditional mathematical knowledge with modern computational methods, emphasizing the lasting impact of Indian mathematical innovations.

Abstract: Although categorized as a pseudoscience, astrology shares many concepts and techniques with mathematics. In this essay, the significance of mathematics is examined with particular attention to the division of the zodiac, the geometric relationships between planets, and the computation of celestial positions. plays in the practice of astrology. To create astrological charts, analyze aspects, and determine when events like eclipses will occur, basic mathematical techniques like spherical geometry, trigonometry, and time conversions are essential. Mathematical models have been used in astrology since ancient times, and astronomical computations form the basis of astrological forecasts. Modern computational approaches employ sophisticated algorithms and software to increase the accuracy and efficiency of these computations. The mathematical foundations of astrology are examined in this work, with a focus on its historical significance and ongoing relevance in the evolution of horoscopes and astrological interpretations. Knowing the mathematical underpinnings of astrology enables this study to emphasize the continuous influence of mathematics on astrological activities by shedding light on the intricate connection between mathematical modeling and celestial observation.

Abstract: In recent times, blockchain technology has emerged as revolutionary technology in various fields, especially in financial and capital markets. The main objective of this research is to analyze the changes in efficiency due to the mathematical concepts of how to improve the performance of the stock market. Three main mathematical models have been used in this study - Market Efficiency Model (Market Efficiancy Model), cost difference models (Cost Differential Model), and risk display models (Risk Exposure Model). These models consider the proportion of transactions, instability, duration of transaction and the relationship between the traditional and blockchain-based system. To support these theories, realistic projects in India and internationally have been included. In India, the background of the National Stock Exchange Blockchain Sandbox Project, CDSL's registration management, and the Reserve Bank of India has been investigated by the background of bond transactions. It has also studied international activities like Nasdaq Linq, Australian Securities Exchange (ASX) and JP Morgan Onyx International activities like this have been studied. In addition, blockchain limits have also been thoroughly analyzed-for example, scalability problems, energy consumption (especially in proof -of-work systems), errors in smart contracts, legal uncertainty and technical complications in combination with conventional systems. These problems have been considered theoretically (such as quaching theory, energy consumption models, game theory, etc.).

Abstract: Voice recognition for gender has come a prominent area of study in machine literacy and speech processing. Dimorphism, or the clear physiological and aural distinctions between man and woman voices, is a point of mortal voices that allows automated systems to determine gender grounded on oral traits like pitch, frequency, accentuation, and speech rate. This study investigates how aural features taken from recorded speech can be used to classify gender using machine literacy algorithms. The delicacy and effectiveness of several bracket algorithms are compared through perpetration and evaluation. According to the analysis, woman voices have slightly advanced frequentness than man voices. Mean frequency of man and woman voice is thick between 0.15- 0.20.

Abstract: Promoting employees is an essential procedure in organizational frameworks that directly affects motivation, productivity, and retention of the workforce. This research investigates data-centric approaches for forecasting employee advancements through classification methods. A collection of 1,000 employee records was examined, featuring variables with 12 like education, age, training scores, last year’s rating, and department. Following preprocessing to manage absent values and encode categorical variables, models such as Logistic Regression, Decision Tree, K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Gaussian Naive Bayes (GNB) were created and assessed. Performance metrics including accuracy, precision, recall, specificity, and F1 score were utilized to evaluate model results. GNB proved to be the best model, achieving an accuracy of 83% on the test data, demonstrating resilience despite class imbalance. The study finds that statistical learning methods can greatly assist human resource departments in making informed, fair, and efficient decisions regarding promotions.

Abstract: Linear Programming (LP) is a mathematical method to optimize results under certain constraints. With the improvement of computational aids, LP problems have become easy to solve. This study compares two well-known aids: Python's PULP Library and the TORA software. Applying a new LP problem, both aids are compared for accuracy, usability and application for various user groups. The research aims to guide students, teachers and practitioners on choosing the best tool depending on their requirements.

Abstract: Customer churn poses a significant concern for the telecom industry, as it directly affects both revenue generation and the efficiency of operations. To better understand and address this issue, the present analysis applies survival analysis methods to study customer tenure and the likelihood of churn. Specifically, the Kaplan-Meier estimator is utilized to estimate the survival function of telecom customers over time, while the Cox Proportional Hazards model is used to assess the influence of various customer attributes on the risk of churn. The study highlights that several customer-related factors play a crucial role in determining the probability of churn. Among these, the type of contract (e.g., month-to-month vs. long-term), mode of payment (e.g., electronic check, credit card), and access to additional services (like internet or tech support) emerged as statistically significant determinants. For instance, customers on short-term contracts or using certain payment methods exhibited higher churn probabilities compared to those with long-term commitments or bundled services. The findings emphasize the importance for telecom companies to tailor their retention strategies by focusing on at-risk customer segments. By understanding the survival patterns and the variables most strongly associated with early churn, service providers can design targeted interventions—such as loyalty programs, contract incentives, or personalized communication—to extend customer relationships and improve overall Customer Lifetime Value (CLV). Ultimately, this evidence-based approach can support telecom firms in minimizing customer loss and maintaining long-term profitability.

Abstract: This study aims to develop a predictive model to assess client responses to bank marketing campaigns. Using an open-source dataset derived from a Portuguese bank’s marketing efforts and hosted on Kaggle, we apply various classification algorithms including Logistic Regression, Random Forest, and LightGBM. The study involves thorough preprocessing, feature engineering, and model evaluation using ROC-AUC and F1 metrics. The best performing model achieved an ROC-AUC of approximately 0.80 using LightGBM, with SHAP analysis revealing the most influential factors.

Abstract: Geometric Deep Learning (GDL) extends traditional neural network paradigms to non-Euclidean data structures, enabling the effective processing of data that lies on manifolds or graphs. Among GDL techniques, Graph Neural Networks (GNNs) have emerged as powerful tools for modelling relational data by leveraging principles from graph theory and algebraic topology. This paper explores GNNs through the lens of mathematics, focusing on how geometric and topological insights drive the architecture and functionality of these networks. By framing GNNs in terms of graph signal processing and spectral theory, we illuminate how GNNs capture dependencies across nodes and edges, offering a structured approach to learning on graph-structured data. We further examine the theoretical underpinnings that make GNNs particularly suited for applications in social networks, molecular biology, and recommendation systems. In doing so, this study provides a mathematical perspective on the capabilities and limitations of GNNs, underscoring the role of invariance, equivariance, and generalization within graph-based learning models.

Abstract: Matrix factorization techniques have emerged as powerful tools in machine learning, particularly for their efficacy in dimensionality reduction and recommender systems. This paper explores various matrix factorization methods, including Singular Value Decomposition (SVD), Non-negative Matrix Factorization (NMF), and Alternating Least Squares (ALS), highlighting their mathematical foundations and computational frameworks. We discuss the significance of these techniques in reducing the dimensionality of large datasets, enabling efficient data representation and storage while preserving essential information. Furthermore, the application of matrix factorization in recommender systems is examined, illustrating how it facilitates personalized recommendations by uncovering latent user-item interactions. Through comparative analysis and case studies, we demonstrate the effectiveness of these methods in addressing challenges such as sparsity and scalability in recommendation tasks. The paper concludes by identifying future directions for research, emphasizing the integration of matrix factorization with deep learning approaches to enhance model performance and adaptability in dynamic environments.

Abstract - A key component in current network administration is remote access to network devices, which helps effective management and troubleshooting. TELNET and Secure Shell (SSH), two of the protocols that provide this kind of access, are widely used for their functions in router setup and upkeep. The security implications of remote router access via TELNET and SSH are critically examined in this work. Due to its early architecture and lack of encryption, TELNET makes network traffic extremely vulnerable to eavesdropping in modern settings. On the other hand, SSH improves the secrecy and integrity of data by providing secure communication channels and strong authentication.

Abstract: The goal of this research paper is to look into how well the Newton-Raphson method works for solving transcendental equations using two different computer programs, Python and Maxima. The Newton-Raphson method converges faster than the bisection method and Regular Falsi method. This is why this iterative method is often used to find the roots of functions with real values. We test the algorithm's speed, accuracy, and ease of use on two different platforms: Maxima, a computer algebra system and Python, a general-purpose programming language with strong numerical libraries. The study gives error analysis, convergence behavior and step-by-step instructions for solving a few selected nonlinear equations. The results show that both tools are good for analyzing numbers, but they have different advantages when it comes to computational control, flexibility, visualization and syntax simplicity. This comparative analysis helps scientists, engineers, teachers and students choose the right tools for solving mathematics problems in science and engineering.

Abstract: Artificial Intelligence (AI) has evolved from a theoretical concept into a transformative force that is actively reshaping modern society. No longer confined to research laboratories or speculative fiction, AI is now embedded in our daily routines—ranging from voice assistants like Siri and Alexa, to complex medical diagnostic tools, self-driving vehicles, recommendation systems, and smart city infrastructure. Its growing presence has made it both an indispensable innovation and a subject of intense social, ethical, and political debate. This research paper aims to explore the intricate relationship between AI and society, investigating how these technologies are impacting various sectors while also highlighting the risks and challenges they introduce. AI offers considerable promise across domains such as healthcare, where it aids in early disease detection and personalized treatment; education, where it enables adaptive learning platforms; agriculture, through smart irrigation and crop monitoring; and public administration, by streamlining governance and improving citizen services. These applications enhance decision-making, increase efficiency, and improve quality of life. However, the widespread integration of AI also raises significant ethical and societal questions. As machines begin to replicate or even outperform human decision-making, concerns emerge around job automation, the erosion of privacy, algorithmic bias, and the opacity of AI decision systems. For example, automated hiring tools may unintentionally discriminate against certain groups due to biased training data, while AI-powered surveillance systems can compromise individual freedoms. Furthermore, the uneven global access to AI technology risks deepening the divide between developed and developing nations. This paper adopts a multidisciplinary and global approach by reviewing existing literature, government policy frameworks, and real-world case studies to assess the double-edged nature of AI's influence. By analysing both the benefits and the harms, the research emphasizes the urgent need for robust governance frameworks, inclusive policy-making, and ethical guidelines. It argues that without meaningful regulation and a commitment to human-cantered design, the risks associated with AI could outweigh its benefits—especially for vulnerable populations. The study also offers forward-looking recommendations for various stakeholders, including policymakers, AI developers, educators, and civil society. These include implementing transparent algorithms, enhancing public understanding of AI, promoting global collaboration on AI ethics, and ensuring fair access to AI-driven tools and services. Above all, it emphasizes that technology must remain a means to empower humanity rather than dominate it. In conclusion, this paper presents a balanced evaluation of Artificial Intelligence's societal implications, urging responsible innovation to harness AI’s full potential while safeguarding human dignity and social justice. As AI continues to evolve, its trajectory must be shaped not only by what is technologically possible but also by what is ethically and socially desirable.

Abstract: Forecasting of precious metal prices accurately is of crucial importance of an informed financial decision-making, robust risk mitigation and strategic asset allocation. This study represents a comparative analysis of time series forecasting methodologies including — Autoregressive Integrated Moving Average (ARIMA), Multilayer Perceptron (MLP), and Support Vector Regression (SVR) applied to the monthly historical datasets of gold and silver prices. These datasets were generated using OpenAI’s ChatGPT for academic purposes. These datasets are simulated and do not directly reflect real-world market data unless otherwise data is validated.  Each of the models is evaluated over a 24-month out-of-sample forecasting horizon using rigorous statistical metrics, including Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). The empirical findings underscore the comparative advantages of data-driven machine learning approaches, particularly in capturing nonlinear and volatile dynamics, with MLP and SVR outperforming ARIMA in most scenarios. These results emphasize the increasing relevance of advanced machine learning techniques in financial time series modelling.

Abstract: In recent days children’s safety is a very big problem for parents and society too! Our child is not feeling safe while traveling. Many incidences occurred where school drivers or other peoples in buses abused children. So, to monitor child’s boarding on bus and drop off from bus we have proposed here real time school bus transport monitoring system. This system uses RFID, Face recognition, GSM and GPS modules embedded with microcontroller. While bus is moving, real time streaming of video is made available to parents from the bus. The model proposed here is very useful in safety concerns of students as well as women’s if used in school buses and company cabs.

Abstract: Insurance fraud is a major problem that threatens both the stability and fairness of insurance systems. This study explores how machine learning techniques—such as Logistic Regression, Decision Trees, Random Forest, and XGBoost—can be applied to identify fraudulent auto insurance claims. The models obtain great accuracy, precision, recall, and F1-score, demonstrating their capacity to distinguish between false and legitimate claims. The performance of the models is further enhanced and improved prediction accuracy is ensured by the use of advanced approaches like feature selection and hyperparameter tuning. Overall, by offering a thorough review of machine learning algorithms and their use in identifying fraudulent claims, this project makes a contribution to the field of auto insurance fraud detection. Insurance businesses can use the created models and procedures to improve their fraud detection processes, reduce financial risks, and safeguard their operations from fraudulent activity Using a real-world dataset from Kaggle, we applied preprocessing techniques, feature selection via Recursive Feature Elimination, and data balancing through SMOTE. Out of all the models tested, XGBoost showed the highest performance, achieving an accuracy of 89% and an F1-score of 87%. The paper highlights the effectiveness of AI-driven detection systems in minimizing financial loss, improving risk management, and ensuring fairness in insurance systems.

Abstract: Generative Adversarial Network is powerful tools for creating new and realistic data to help in to improve machine learning models, specifically when there’s not enough labeled data. It has two parts: Generator-which create a fake data and Discriminator-which tries to tell real data from fake. Through the continuous competition, generator gradually learns to create increasingly realistic data. This paper looks at how GANs can be used to make more data, helping with problems like unbalanced classes and over fitting. It also explains how newer types of GANs, such as Conditional and Wasserstein, increase training stability and enhance the caliber of the data they produce. We also share real-world examples of how GANs are used in different areas, like identifying images analyzing medical scans, and understanding language. These examples show that using GANs to create extra data can really help improve machine learning results. In the final part of paper, we talk about some of the problems that still need to be solved and what the future might look like for this technology. We also explain why it’s important to use both real and fake data carefully, so that models stay accurate and works well.

In public areas, QR codes are being utilized more and more for information sharing, marketing, and payment. But because of their ease of use and user confidence, they are open to abuse, such as phishing, malware distribution, and illegal data access. In order to examine the effects and create defenses, this study replicates actual QR code-based attack scenarios in controlled ethical hacking environments. In order to propose defenses strategies like QR code validation, user awareness, and embedded link scanning, the paper investigates how to set up safe lab conditions for QR-based social engineering, redirection attacks, and malicious payload.