Authors - Twinkle K. Patel, Ankit Bhavsar, Arpit Jain Abstract - Automatic Speech Recognition (ASR) systems have become essential tools for the communication in natural human-computer communication across various languages. The growing need for effective communication in public and private sector has increased the need of Natural Language Processing. Gujarat is one of the prominent state in India where most of the official communication takes place in Gujarati language. ASR for Gujarati language has its own importance due to the high usage of Gujarati language. This paper, highlights the emergence of ASR for Gujarati language from the early days of creating ASR commercially to more modern advancement approaches such as deep learning, end-to-end (E2E), and various other methods. Traditional ASR systems utilized statistical methods, most commonly Hidden Markov Models (HMM) and Gaussian Mixture Model (GMM) which often reported low accuracy. With the emergence of deep learning models, including CNN, BiLSTM, and transformer-based models such as Wav2Vec 2.0 and XLSR-53, the recognition accuracy has significantly increased, especially in noisy and spontaneous speech contexts. The paper also provide the comparative study of the work done by various researchers in the field of Gujarati ASR, highlighting that the use of spell correctors as well as hybrid feature extraction methods have reduced phonetic ambiguities and diacritic errors in Gujarati language. The study strongly emphasizes the need for morphological analysis and contextual modeling to adequately address the complexity in the homophones and diacritics of Gujarati language. The paper also reviews the various methods applied to Gujarati ASR and their reported Word Error Rates.
Authors - Prathilothamai M, Nitin Ravi, Rishi Pradeepkumar, Sajith Rajan P, Anurag Nagilla Abstract - Code analysis ensures software quality, readability, and maintenance. Traditional methods such as Static Analysis Tools (SAT) and SZZ algorithms recognize errors and analyze contributions based on predefined rules and historical trends. However, complex code semantics are difficult to understand and often lead to overlooked issues and limited feedback. Despite its benefits, LLM is not practical in some business environments, facing challenges such as high computing costs, slower processing, and data protection concerns. Hybrid models may provide scalable and inexpensive solutions for large enterprise projects. This study contrasts SATs and LLMs, focusing on capabilities to detect bugs and maintain code quality, by investigating hybrid models that fuse the contextual strength of LLMs with the efficiency of SATs. In evaluations with 30 participants, the hybrid model consistently received the highest scores across accuracy, clarity, relevance, and usability (average ratings >8.5/10), outperforming both SATs and AI-alone approaches. Hybrid models may thus provide a scalable and cost-effective solution for problems arising in large-scale corporate enterprise projects. This study supports the use of case studies and benchmarks to assess the effectiveness, cost, and scalability of these various approaches. The findings aim to give practical suggestions for improving traditional code review processes, helping balance rule-based tools with AI-powered insights to enhance software development workflows.
Authors - T. A. Alka, M. Suresh Abstract - The study aims to examine the link between social innovation (SI) and sustainable entrepreneurship (SE). The major focus is to explore how SI is promoted by the SE and what the Sustainable Development Goals (SDGs) are covered by the sustainable entrepreneurship through promoting social innovation identified through a research gap by adopting the TCCM framework. The study is a systematic literature review on the publications in the Scopus database identified through keywords relevant to the theme by adopting the SPAR-4-SLR protocol. The result indicates that the sustainable entrepreneurs promoting SI innovation through innovative business models ensure social well-being. The SI is not only the technological but also non-technological aspects, which are responsible and impactful, having innovative approaches, social practices, and processes. The SEs are developing business models that are socially and sustainability purpose-driven, offering financial advantage and ensuring the social cohesion and the empowerment of the community through their participatory approaches and ensuring the local level stakeholders’ engagement. SE develops partnerships and collaboration among the different actors, which will create knowledge sharing and collaborative innovation. While mapping of SDGs to SI by SEs, the study identifies that the major SDGs covered are: SDG 1, SDG 4, SDG 5, SDG 8, SDG 9, SDG 10, etc. The study offers implications to practitioners, policymakers for actionable strategies to enhance the SI-driven SE ecosystem. This systematic review lacks empirical findings and relies purely on the published works. The limitations will open future research scope avenues by covering more databases, keywords, etc.
Authors - Shaveta Thakral, JyotiVerma, Pratima Manhas, Shankar Amalraj, Neha Shahare, Monali Tingane, Aparna Mahesh Torade Abstract - Urbanization and increasing housing density have necessitated innovative solutions to address both space constraints and accessibility for individuals with limited mobility. This paper presents a novel concept of an integrated lift-toilet system, designed specifically for elderly and disabled individuals living in multi-story buildings. By merging the functionalities of vertical transportation and sanitary facilities, the proposed system aims to provide dignity, convenience, and emergency usability, particularly in high-rise residences and healthcare environments. The research encompasses a multidisciplinary approach, including mechanical design, sanitary engineering, user-centric ergo-nomics, and real-world application assessments.
Authors - Jyotika R. Yadav, Arpit A. Jain, Ankit Bhavsar Abstract - Efficient traffic management is a major focus in smart city projects. Intelligent Transportation Systems (ITS) are revolutionize the way transportation networks are managed, to improve safety, efficiency, and sustainability. Machine learning (ML) has been developed as a powerful means for advancing ITS by enabling data-driven decision-making, real-time traffic management, predictive analytics, and automation. This paper offers a thorough summary of the current trends and advancements in the utilization of ML techniques in ITS. It explores key areas such as traffic flow prediction, autonomous vehicles, congestion management, smart mobility solutions, and road anomaly detection.
Authors - Saranya P., Rejikumar G. Abstract - This study employs a systematic literature review (SLR) to investigate the changing landscape of smart labelling in the food industry. As Customer demand for transparency, traceability, and ethical consumption increases, conventional food labels are evolving into interactive communication instruments. Smart labels facilitated by technologies such as QR codes, RFID, and intelligent packaging are now integral to real time customer interaction and business decision making. Despite increasing scholarly interest, a thorough delineation of research trends, deficiencies, and prospective directions has been absent. To tackle this, adoption of SPAR-4-SLR protocol was done to examine 43 peer reviewed papers (2015-2025) from Scopus utilizing Biblioshiny and Vosviewer. The investigation identified few predominant topics like food safety, customer behaviors, and technological integration while also highlighting under discovered topics like RFID and packaging designs. Result indicated a shift from a compliance orientated labeling to an interactive digital approach that foster consumer trust. This review provides essential in-sights for scholars, legislators, and industry executives seeking to create data-driven, customer-focused labelling systems in the digital era.
Authors - Abhishek Passwan, Srushti Karki, B M Bandgar Abstract - Accurately predicting stock prices remains a challenging task, primarily due to the volatile, dynamic, and non-linear nature of financial markets. This study introduces a robust deep learning-based framework designed to forecast the closing prices of companies listed on the Nifty 50 index. We explore and compare the predictive performance of several state-of-the-art neural network architectures, including Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Recurrent Neural Networks (RNN), and Convolutional Neural Networks (CNN), alongside two novel hybrid models: StockAugNet-f and StockAugNet-c. Historical stock data were retrieved from Yahoo Finance and pre-processed using Min-Max normalization. To effectively model temporal dependencies, we applied a sliding window technique for sequence construction during training. Model performance was assessed using key evaluation metrics: Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). To enhance usability and support real-time forecasting, we developed an interactive Streamlit-based application, allowing users to dynamically visualize model outputs. Experimental findings indicate that the GRU and RNN model give better accuracy over the hybrid models accuracy and resilience for the NIFTY 50 index stocks.
Authors - Kumar Gaurav, Sudhir Agarmore, Akshay Kumar, Akshat Jha, Edidiong Akpabio, Chandan Kumar Abstract - Network architecture is examined in-depth in terms of the structural design or the underlying protocols that help in the facilitation of data transmission between networks. The paper discusses the classic models of network architecture, which include P2P, client-server, and hybrid models, and outlines the latest advancements in cloud computing, edge computing, and IoT. The paper further discusses the different types of network topologies and other devices, such as the OSI and TCP/IP models, in the shaping of data communication. This also encompasses new emerging trends, such as 5G, IoT, quantum networking, and zero-trust security, among many others. These will detail discussions of the concepts involved as well as their impacts on network protocols and architecture. As demand for faster, more secure, and scalable networks grows, the integration of these modern technologies opens the way to new opportunities and challenges of future network infrastructure.
Authors - Janani Vijayan, Bala Subramanian R Abstract - This study explores the growing body of research on leader’s gratitude, emphasizing its significance in organizational settings. Gratitude, a complex and multifaceted positive emotion, plays a crucial role in enhancing interpersonal relationships, promoting well-being, and fostering prosocial behavior. The review categorizes gratitude into trait, state, and collective forms, examining their unique characteristics and implications. It highlights key theoretical frameworks such as the Find-Remind-and-Bind Theory, Broaden-and-Build Theory, and Cognitive Emotion Theory to explain how gratitude functions within leadership dynamics. The review further delves into the antecedents of leader’s gratitude, including team error management, feedback, and appreciation programs and its consequences, such as enhanced social intelligence, team innovation, improved leader-member relationships, and increased organizational commitment. Moreover, the paper discusses the broader social impacts of leader gratitude, showing how it fosters a positive work culture. Identified limitations include the scarcity of causal studies and underrepresentation of cultural factors. Future research directions are proposed, such as exploring gender differences in gratitude expression, cultural variations, and the effectiveness of different gratitude expressions. Overall, the review underscores that leader’s gratitude is a powerful yet underutilized tool in cultivating effective leadership and healthy organizational environments.
Authors - S. Kirthica, K. Subha, S. Arul Murugan, V. Deepak Kumar, M. Dhivagar, S. Kanishka Abstract - In contemporary society, effective grievance redressal is vital for transparency, accountability, and public trust. Traditional systems often face issues like manual processing, delays, and poor user engagement. To overcome these challenges, we propose VOICE, an AI-driven grievance management platform that automates the complaint lifecycle—from classification and similarity detection to assignment and tracking. Leveraging natural language processing (NLP), speech-to-text, and automated prioritization, VOICE reduces human intervention and improves response times. It also incorporates community-driven upvoting, real-time notifications, and performance monitoring to enhance user engagement and system transparency. Designed for scalability and ease of use, VOICE can be deployed across institutions, communities, and national systems. By combining automation, intelligence, and user-centric features, VOICE represents a significant advancement in modern grievance redressal.
Authors - And Julie D. Rosal Abstract - This study checks on emerging research trends and conducts a bibliographic analysis on smart waste bins as foundational elements for cost-effective, automated waste management solutions. The escalating urbanization and environmental pressures necessitate advanced waste management technologies, where smart trash bins equipped with IoT (Internet of Things), AI (Artificial Intelligence), and robotics offer transformative solutions such as automated segregation, real-time monitoring, and optimized collection routes. Employing a systematic literature review facilitated by Scopus AI, the study identifies consistent, rising, and novel themes in smart bin research. IoT-enabled systems consistently emerged, underscoring their role in operational efficiency and sustainability. Rising trends prominently featured AI integration, robotic automation, and advanced waste segregation methods, indicating a clear shift towards intelligent automation. Security and privacy issues constitute novel themes, reflecting the growing importance of ensuring trust and reliability in technology adoption. The bibliometric analysis revealed significant global research contributions with notable peaks aligning with heightened global sustainability goals. Additionally, the study proposes a conceptual prototype integrating sensors, AI-driven categorization, and autonomous collection mechanisms, validated through simulation testing. Findings demonstrated improved operational efficiency, higher recycling accuracy, and substantial cost savings. Addressing integration challenges, infrastructure compatibility, and public acceptance, the study provides critical insights for urban planners, policymakers, and stake-holders. This work significantly contributes to the discourse on sustainability and smart city initiatives, highlighting technological innovation as pivotal in solving contemporary environmental management issues.
Authors - Akheel Mohammed, Sameera Khanam, Ayesha, Mohammed Shah Alam Rasool Khan Abstract - The proliferation of cloud computing presents a significant challenge for enterprises: entrusting mission-critical data to remote service providers. The widespread adoption of cloud services is fundamentally hampered by a lack of robust and reliable trust evaluation mechanisms. To address this critical obstacle, this paper introduces a novel reputation-based trust model designed to accurately assess the trustworthiness of Cloud Service Providers (CSPs). Our proposed model uniquely integrates three key metrics: direct customer feedback, historical server rejection rates, and real-time server workload. A specialized trust evaluation algorithm processes these inputs to generate a comprehensive reputation score. Experimental results validate the efficacy of our approach, demonstrating a more efficient and accurate evaluation of CSP trustworthiness compared to existing models.
Authors - Ashwini Matange, Jibi Abraham Abstract - Major component of research is searching for relevant publications in the domain of research from varied sources, exploring them, organizing them and finding interconnections between them. Different platforms ranging from Google Scholar, PubMed, IEEE Xplore, ScienceDirect, Springer are to be visited and explored to find research papers in the relevant field of the researchers. Research Rabbit is a free to use literature mapping tool that provides a one stop venue to locate papers of interest effortlessly with effective visualizations and smooth navigation.
Authors - Seema Shrawne, Aishwarya Jadhav, Saniya Gupte, Sayali Panch, Sakshi Thombre, Vaibhav Dhore, Vijay Sambhe Abstract - Accurately detecting and counting fruits is crucial for improving yield estimates in Precision Agriculture. However, challenges such as varying fruit types, overlapping objects, and cluttered backgrounds make this task difficult. In this study, we use RetinaNet, a deep learning model known for detecting dense objects, to address these issues. By replacing its ResNet50 backbone with SEResNet18, we improve both accuracy and speed, increasing mAP by 1.65%at IoU 0.5 and 1.79%at IoU 0.7. We also fine-tune the model using Differential Evolution to optimize anchors, further boosting mAP by 0.61% and 7.3% at IoU 0.5 and 0.7, respectively. Our enhanced model performs better on KFuji apples, mangoes, and holly fruits, while the default model slightly outperforms ours on strawberries and oranges at higher IoU thresholds.
Authors - Meeradevi, Maria Rufina P, Prathik B, C S Parthasarathy, Janya V, Neeha Ritvika Mandava Abstract - PathoVision is an AI-powered framework that enhances cancer diagnosis by integrating histopathology images with clinical text using multimodal learning. Unlike traditional systems focused solely on image analysis, it employs the PLIP model to generate combined visual-text embeddings for more accurate and contextual understanding. A custom decoder pro- cesses these embeddings to classify cancer subtypes, while Grad- CAM provides interpretable heatmaps, ensuring transparency in decision-making. Designed for use in clinical diagnostics, research, and education, PathoVision addresses the limitations of manual pathology—such as subjectivity and time constraints—by offering a reliable, interpretable, and accessible solution. This framework represents a significant advancement in computational pathology, blending modern AI with traditional diagnostics to support expert pathologists and improve patient outcomes.
Authors - And Julie D. Rosal, David Livingstone T. Ramos, Randy Joy M. Ventayen Abstract - This study explores the quality assurance (QA) practices of Pangasinan State University (PSU) from 2016 to 2025, emphasizing key strategies that enhanced its national and global reputation. Using a multi-method approach—including literature review, surveys, and interviews, examines stake-holder perceptions, institutional challenges, and achievements in quality assurance recognition, and improved international rankings. The findings highlight the importance of strong leadership, continuous improvement, faculty development, and stakeholder engagement in fostering a culture of quality. These elements have strengthened PSU’s quality management system and its responsive-ness to academic and institutional demands. As a forward strategy, the study presents a five-year roadmap for 2026–2030 focused on sustaining quality through digital transformation, global benchmarking, research-driven planning, and performance-based evaluation. This roadmap is intended to solidify PSU’s status as a globally competitive university and offers practical insights for other state universities aiming to institutionalize QA and improve their academic standing at both national and international levels.
Authors - David Livingstone T. Ramos Abstract - Quality assurance (QA) in higher education is a critical mechanism for promoting academic excellence, institutional ac-countability, and societal impact. State Universities and Colleges (SUCs) in the Philippines serve as gateways to education for under-served regions but face significant challenges in maintaining and im-proving quality due to the absence of standardized frameworks, reliance on manual processes, and limited technological integration. These issues hinder their efficiency, compliance with accreditation standards, and ability to respond to evolving educational demands. This study examines the role of a Quality Assurance Information System (QAIS) as a solution for addressing these challenges. Employing a literature review methodology, it evaluates current QA practices in SUCs and identifies gaps, such as fragmented processes, inconsistent standards, and insufficient data management capabilities. The research emphasizes the potential of QAIS to streamline processes, automate routine tasks, and provide real-time data for evidence-based decision-making. Key components of an effective QAIS include modular design for scalability, compliance with international standards, user-friendly interfaces, and robust analytics for monitoring and continuous improvement. By implementing QAIS, SUCs can align their practices with global benchmarks, enhance stakeholder engagement, and improve institutional performance. The study also highlights challenges in adopting QAIS, including resource limitations and resistance to change, and proposes strategies to mitigate these barriers. The findings offer actionable recommendations for SUCs to implement and scale QAIS effectively, aligning with national and international quality standards while addressing local needs. By leveraging QAIS, SUCs can enhance their competitive-ness, improve educational outcomes, and contribute to broader national development goals.
Authors - Pravitha N R, Sumedha N Varma Abstract - This study reflects the pivotal role of financial technology (Fin-tech) in reshaping the IT and Corporate sectors while playing a role in achieving sustainable development. Fintech has evolved into a significant role as a game changer, driving innovation in financial contribution through digital platforms, blockchain technology, artificial technology, and big data analytics. In the IT sector, Fintech facilitates secure, efficient transactions and enhances digital infrastructure, leading to improved service delivery and client engagement. Within the corporate domain, it streamlines operations, increases financial inclusion, and supports strategic decision-making. In addition, the integration of Fintech services with principles of sustainability is enhancing green finance and promoting efficient digital solutions. This paper studies the factors affecting the adoption of fin-tech services and how sustainable development goals are achieved by adopting fin-tech services. By using regression analysis, the study expresses that Digital Financial Literacy and Perceived Performance Expectancy have a positive impact on the Behavioural Intention in adopting fintech services. The analysis further states that the actual use of fin-tech services plays a vital role in promoting sustainable development.
Authors - Sasini Gandhari Dabare, Indra Mahakalanda, Prasanga Jayathunga, Nathali Fernando Abstract - Media Mix Modeling (mMM) is a data-driven analytical approach that quantifies the influence of multiple advertising channels on sales performance while adjusting for external factors such as seasonality and economic conditions. This study explores machine learning-based mMM to predict weekly sales performance in the Sri Lankan home appliances industry by combining ad spending variables with external factors such as exchange rates and holidays. The study applies time series decomposition, ad stock transformation, and data normalization to capture delayed and nonlinear media effects. TimeSeriesSplit cross-validation was used to test four predictive models: Linear Regression, XGBoost Regression, Support Vector Regression (SVR), and Bayesian Ridge Regression, along with key metrics for performance (MSE, RMSE, R² Score, MAPE). The tuned XGBoost model had the best accuracy (R² = 0.9145; MAPE = 2.87%), making it most useful for predicting. The findings demonstrate the importance of machine learning-enhanced MMM as an effective strategy for optimizing media spending and increasing advertising efficiency in competitive, budget-constrained markets.
Authors - Hera Khairunnisa, Nayla Nandhita Nuril Hadi, Surya Anugrah, Rochma Sudiati, Windy Permata Suyono Abstract - First objective of this study is evaluating MSMEs tax compliance with Strengths, Weaknesses, Opportunities, and Threats (SWOT) analysis. Second objective of this study is evaluating the role of Core Tax Administration System (CTAS) for MSME. The study uses a descriptive qualitative approach, with semi-structured interviews with three tax consultants with extensive experience in MSME taxation. This study also employs the literature review approach. From the interview, this study shows that MSMEs typically have straightforward cost structures given their simple business operations. Government support, such as the reduction of the final tax rate for MSMEs from 1% to 0.5%, is viewed as an opportunity in taxation. However, the study highlights ongoing challenges in MSME operators' tax competence, particularly their understanding of recent tax regulations and literacy in tax matters. This study also shows that the implementation of the Core Tax System serves as a strategic advancement in enhancing tax administration for MSMEs in Indonesia. By leveraging digital integration and automation, the system aims to promote greater efficiency, transparency, and ac-countability in tax processes.
Authors - Samiksha Andeo, Supriya Narad Abstract - How we've been interacting with technology has altered with the presence of chatbots, which have now become a part of our digital society. These artificially intelligent virtual personal assistants are being applied on a vast scale in customer services, health care, and education sectors since they are capable of answering the customers, addressing their queries, and directing them in real-time. In relation to their application in healthcare, in this research study, the his-tory of chatbots is discussed, the technology employed by them, their real-world applications, and the problems faced by them. Modern-day chatbots can understand human language, learn from past conversations, and communicate in a customized way by employing Natural Language Processing (NLP), Machine Learning (ML), and Artificial Intelligence (AI). They may be anything from AI-based chatbots able to engage in meaningful, dynamic conversations to rule-based assistants that utilize pre-defined scripts. With the advent of technology, chatbots are increasingly being applied in the field of health care such as patient assistance, mental assistance, and diet and nutrition guidance, all of which enhance access to health care services. However, despite all their benefits, their potential is also hindered by limitations such as privacy, ethical concerns, and the inability to understand human feelings. As the future of deep learning, IoT connectivity, and more interactive AI is on the horizon, chatbots appear to have a very bright future. The purpose of this research is to give an authoritative and illuminating perspective on how chatbots are changing various businesses and the future of AI communication.
Authors - Sadman Hafiz, Md. Ataur Rahman, Tahmid Zamee, Md Sacklain Hossain, Marufa Akter, Ahmad Mostofa Kamal, Mahady Hasan, M. Rokonuzzaman, Farzana Sadia Abstract - The organic agriculture sector in Bangladesh faces significant challenges due to the absence of a national organic product certification system. As a result, local producers are often forced to rely on expensive international certification agencies, creating barriers to market entry, increasing costs and limiting consumer trust. To solve this, our proposed system is a blockchain-enabled, AI-powered web platform designed to establish a secure, transparent, and efficient organic product certification system tailored for the Bangladeshi context. A key innovation is its hash-based file verification to minimize gas fees to store data on the chain. Farmers or producers submitted product data like images, videos, and documents, which are encrypted (AES-256) and stored locally, while cryptographic hashes (SHA-256) are saved on-chain. Any alteration in a file changes its hash, and QR-code scanning will instantly flag tampering, ensuring product integrity. It also integrates tokenized payment systems using ERC-20 tokens alongside SSLCOMMERZ payment gateway, ensuring secure, traceable, and accessible transactions for all stakeholders. Key features include smart contract automation, real-time certification tracking, a decentralized dispute resolution mechanism, and QR code-based traceability for consumers. This project, therefore, aims to democratize organic product certification, lower operational costs, and build consumer trust, while aligning with global trends in digital food supply chains.
Authors - Wani Tejaswi, Pawar Suvarna Abstract - The number and intensity of car accidents around the world are rising, which makes it even more important to have smart systems that can quickly and correctly process accident pictures. We need more advanced systems to quickly find accidents, label data, and record them. Manual methods take a long time, are prone to mistakes, and are not good enough for real-time apps. This paper tries to solve these problems by suggesting a system that uses advanced machine learning (ML) models to automatically process, label, and report images of car accidents. Convolutional Neural Networks (CNNs) are used for feature extraction, YOLOv8 is used for real-time accident recognition, and Transformer-based models are used for complex multi-object labeling Global Road Accidents Dataset is much better at generalizing models than the Car Crash Dataset and the Road Traffic Accidents Dataset, which are more focused on crash types and vehicle damage. This is because it has more images and a wider range of features, such as location, severity, and weather. It was tested and found that the proposed hybrid model is more accurate than the CNN- based and YOLO-only methods, which got 94.5% and 93.7% accuracy and precision, re-spectively, on the Global Road Accidents Dataset. The combination model also has better accuracy for annotations and fewer fake hits. Comparative research shows that combining Transformer designs with object recognition models makes it easier to understand features and make reports. Findings from this study show that the AI-driven framework can be used to automate accident investigation processes. This is a big step towards smart traffic control and emergency response systems.
Authors - Hasin Mahir, Tahfizul Hasan Zihan, Md. Shirazim Munir, Khondkar Ayaz Rabbani, Rifat Ara Rouf, Ferdows Zahid, Mahady Hasan, Md. Tarek Habib Abstract - Indoor air quality (IAQ) is critical for health, comfort, and cognitive performance in classrooms. Yet, many classrooms in lowermiddle-income countries (e.g., Bangladesh) rely on natural ventilation and lack any warning system for impending air-quality guideline breaches. This study is motivated by the need to maintain healthy learning environments in resource-constrained settings, where students and teachers are often exposed to elevated CO2 and particulate levels. To address this, we present the first comparative evaluation of three forecasting models: Prophet, Random Forest (RF), and Long Short-Term Memory (LSTM), on daily PM2.5 and CO2 time series covering six months of continuous operation in Dhaka, Bangladesh. Our goal is to anticipate IAQ deterioration to enable proactive ventilation or filtration interventions. The nonlinear machine-learning models substantially outperform the Prophet baseline. Random Forest performed best for PM. (RMSE 1.87 μg/m³, R² 0.992), showing its ability to capture complex pollutant dynamics. LSTM excelled at forecasting peak CO2 (RMSE 159.7 ppm, R² 0.693), which is critical for timely interventions. These findings demonstrate the feasibility of accurate, low-cost IAQ forecasting in resource-constrained classrooms and underscore the potential of data-driven forecasting to maintain healthier learning environments.
Authors - Thuy-Vi Thi Ha, Phuoc-Hung Vo, Thanh-Nghi Do Abstract - In this paper, we propose an enhanced deep learning approach for early-stage rice yield prediction by embedding self-attention layers into established deep neural networks (DNNs) such as VGG-16, DenseNet, MobileNet, ResNet, Inception, and Xception. The addition of self-attention significantly improves the models’ ability to capture longrange dependencies and global context, which traditional convolutional layers often fail to represent adequately due to their inherently local receptive fields. We evaluate the proposed method on a dataset of 18,642 RGB images collected from 47 rice field plots spanning over 28 hectares in An Giang and Tra Vinh provinces. The images were acquired using digital cameras, smartphones, and fixed-wing UAVs during the heading stage of rice growth. Experimental results demonstrate that DNNs enhanced with self-attention layers consistently outperform their original fine-tuned counterparts. Furthermore, these hybrid models also achieve higher prediction accuracy compared to Vision Transformers (ViT), highlighting the effectiveness of integrating self-attention into conventional DNN architectures for agricultural yield forecasting.
Authors - Soham Paithankar, Khushi Khedkar, Supriya Narad Abstract - In this research paper, there is a new forecast on the programme that will be able to predict the weather for agriculture. Enhanced by the Internet-connected weather and climate models, earth observation data, and artificial intelligence algorithms, the system returns to farmers accurate, location-based outlook. It does this by factoring preceding day’s temperature data and other factors to determine the needed plant growth, therefore offering comprehensive and precise directions. Furthermore, the alarm systems help in giving the farmers a prior indication of any future possible weather-related incidents that cause a threat to human and their stock’s safety so that measures could be taken accordingly. It also uses real information especially in agriculture sector thus transmits present information to the farmers. The main objective that the system has for farmers is to provide them with information on crops during a favourable time for crop production and the main idea of achieving this is to boost agriculture sustainability and productivity, this is because climatic conditions are very volatile thus would have a great negative impact on farming. Also, the expected audience becomes more diverse and can principally grow, as we would like to consider farmers, who may not even have a basic computer knowledge, empowered by the platform.
Authors - Ankit Shah, Roshni Rawal Abstract - Artificial intelligence-powered learning analytics (LA) shows great results for assisting teachers with the time-consuming task of providing feedback in context to ethical consideration, leads to fairness, maintaining privacy and transparency in feedback. The study investigated two GPT model versions (3.5 and 4) that produce evaluation advice on students' writing assessment based on assignments in context of data science subject. Researchers evaluated GPT generated advice in comparison with human instructions on the basics of effective-ness, readability, and reliability. The paper concludes that both versions could constantly produce more decipherable advice in comparison with human instructions, but GPT-4 performed much better than GPT 3.5 and human instructors. Researchers talked about the unique blend of ethical contemplation and the impact of LLM’s on automated feedback and data analysis provides a perfect match to the LLMs to achieve ethical horizons and set a new height to be achieved for upcoming generations.
Authors - Shalini S, Mamatha A, S. Sheela, Mala B A, Nagaraj M Lutimath, Koustav Biswas Abstract - Sustainable soil management is essential for ensuring food quality, environmental health and climate resilience. This systematic survey examines the evolution and integration of Artificial Intelligence (AI) and Machine Learning (ML) within green agricultural technologies aimed at enhancing soil health. This paper focuses on sensor networks, remote sensing, Internet of Things (IoT), robotics and data-driven decision-support systems to assess their efficiency in precision soil diagnostics, fertility management and resource conservation. While these green AI technologies show promising benefits, improved resource efficiency, reduced environmental footprint, regenerative soil practices challenges persist, including uneven soil data availability, high implementation costs and limited interpretability of models and lack of standardization.
Authors - Aida Ouedraogo/Rakissga, P Justin Kouraogo Abstract - We're interested of the control plane's fault tolerance and the solution we're proposing is proactive thanks to its modularity. Our solution proposes a distributed SDN (Software-Defined Networking) architecture, comprising a controller specifically dedicated to processing packet_in messages and a generic controller in charge of other network management functions. The goal is to prove the effectiveness of this role specialization with regard to performance in message processing and the ability to resist failures.
Authors - Shyamali Thasale, Seema Kedar, Rutuja Khedkar, Kartik Naphade, Prajakta Gaikwad, Vijay Kale Abstract - Sentiment analysis of tourist reviews examines common points of view to help users choose destinations, accommodations, and services. Provides businesses with information on customer satisfaction and helps travelers make better choices based on a variety of common experiences. The proposed method uses a mix of the Skip-Gram models and Continuous Bag of Words in addition to Word2Vec vectorization to transform textual data into numerical values. This method maintains the sentiment of the text and also captures the context and word relationships. Convolutional neural networks boost the accuracy of classification and feature extraction. Along with CNN, models like Random Forest, AdaBoost, LSTM and CNN-BiLSTM have also been used in order to hold a comparison between them. Among these, the CNN model showed the highest accuracy of 96.3 %. This system provides users with specific recommendations for vacation spots, accommodations, and activities.
Authors - Syamanarayanaprabha S S, Abymon A, Aarathy K Sudhakaran Abstract - Digital transactions in India have increased significantly since the National Payments Coperation of India introduced the Unified Payments Interface (UPI) in 2016. This study focuses on Generation X (ages 45-60) in Kerala, a state known for its high level of digital literacy, whereas adoption studies have mostly focused on younger generations like Millennials and Gen Z. The results, which are based on responses from 132 participants, show that security, transaction speed and convenience of use are important element that promote UPI adoption within this demographic. However, technological difficulties and security-related worries still serve as obstacles. To increase UPI acceptance in this group, the study emphasizes the need of more user-friendly application designs, focused digital education initiatives, and calculated interventions.
Authors - Salvatore Vella, Fatima Hussain, Salah Sharieh, Murrium Zaheer, Alex Ferworn Abstract - Artificial Intelligence (AI) using Large Language Models (LLMs) is now used to automate simple knowledge worker tasks such as creating communications and summarizing documents. This paper presents a domain-specific language for creating agents, which we call ContentCreator (CC), to perform common knowledge worker tasks and mimic the behaviour of knowledge workers and their workflows. An interpreter has been built to execute and automate these workflows. In the same way that computer languages have progressed from assembly language to increasingly higher-level languages, CC enables the high-level programming of workflows that utilize large language models. CC is a simple language that non-programmers can use and understand. We use several use cases to demonstrate the ability of ContentCreator to automate workflows across multiple tasks.
Authors - Bryan Atunca, Henry Coico, Edgar Ramos, Araceli Macassi, Donovan Fuqua Abstract - The adoption of digital technologies and intelligent systems is significantly enhancing efficiency and safety across the global agri-food industry. Integrating Industry 4.0 objectives with circular economy principles offers effective solutions to challenges such as resource scarcity and environmental impact. This study proposes a leagile supply chain model—combining lean and agile method-ologies—to optimize mango supply chains worldwide. By leveraging Cyber-Physical Systems (CPS) in postharvest processes such as drying and storage, the model enables remote monitoring and control of key environmental parameters. CPS technologies, through sensors, actuators, and data-driven algorithms, provide real-time feedback, minimize human error, and support precision in decentralized processing units. Findings show a reduction in total supply chain time from 41 to 30 days (26.8%), with efficiency gains of 25% in harvesting and 28.5% in storage. The integration of CPS strengthens circular supply chain practices, reduces losses, and enhances sustainability, profitability, and system resilience across global mango supply chains.
Authors - Abhijit T. Somnathe, Parameshwari M, Sampurnima Pattem, Sridhar N Koka, Shokhida Abdurakhmanova, Pundru Chandra Shaker Reddy Abstract - Gold prices are non-linear, unexpected, volatile, and unregulated, making prediction difficult. Numerous studies have anticipated gold prices because they affect international economic and monetary systems. However, linear relationship studies rarely explain gold price changes. Gold price time series data is unpredictable, nonlinear, and volatile, making prediction difficult. Classical statistics and machine-learning (ML) approaches like Random-Forests, CNNs, and RNNs offer excellent accuracy but limits. A model that combines Temporal-Convolutional-Networks(TCN) with Query (Q) and Keys (K) attention-mechanisms (TCN-QV) is presented to improve gold price forecasts. The model extracts temporal properties from sequence data using stacked dilated causal convolution layers in the TCN architecture. To adapt weight distribution to information features, an attention mechanism is introduced. Finally, dense layers give projected results. This approach predicts Shanghai gold price time-series. The optimised model reduces Mean Absolute Error (MAE) by 5.47% in the least favorable case and 33.69% in the most favorable case across four investigational datasets compared to the baseline model.
Authors - Pooja Anil Patil, Yashodhara Haribhakta Abstract - Named Entity Recognition (NER) is a core task in Natural Language Processing (NLP) with applications in conversational AI, healthcare analytics, and search systems. It enables structured extraction of key entities from text, supporting clinical decision-making and automated diagnosis. While NER resources for high-resource languages like English are widely available, low-resource languages such as Marathi lack domain-specific datasets, particularly in the mental health sector. This work introduces ArogyaMINDNER, a Marathi mental health-specific NER dataset, designed to enhance entity recognition for diseases, symptoms, and treatments. The dataset is manually annotated using BIO tagging, ensuring linguistic accuracy and contextual consistency across formal and conversational text. Annotation policies account for psychiatric terminology nuances, accommodating Marathi’s morphological richness and linguistic diversity. To assess performance, ArogyaMIND-NER is benchmarked using classical machine learning models (SVM, Naïve Bayes, XGBoost, LightGBM, Random Forest), deep learning architectures (CNN, BiLSTM, BiLSTM-CRF), and transformer-based models (mBERT, XLMRoBERTa, IndicBERT, MahaBERT, MahaMarathi-7B). Comparative evaluation reveals MahaMarathi-7B and Marathi-Social-NER as topperforming models, achieving higher precision and recall in disease recognition, while classical models struggle with contextual understanding. This dataset and trained models establish benchmarks for Marathi NLP research, contributing to mental health analytics, clinical applications, and low-resource language processing. The findings advocate for finetuning transformer architectures for domain-specific NER, ensuring better entity identification in Marathi mental health texts. The dataset and models will be made available for further research and development.
Authors - Navin Barde, Deepak S. Sharma Abstract - Good E-Learning system for colleges would ensure student learning to be flexible, accessible, and interactive for being remote learning option in the Post COVID-19 times. The most important aspects are an intuitive and easy to use GUI to easily browse on any device, the interactive content such as video lectures, simulations, quizzes, forum to improve the engagement. Customized learning paths, progress tracking, grade management, and instant communication features enhance learning. We take security very seriously, we encrypt all of your data while storing it and ensure that students privacy is maintained with strong user authentication. Laying particular emphasis on usability, interactivity, personalization, and security, such systems create a convincing and flexible learning environment, which encourages life long learning.
Authors - Mohit Menghnani Abstract - The increasing expenses and shortage of qualified developers have driven many to look to software development with an eye toward the incorporation of Artificial Intelligence (AI). Full-stack development facilitates the web application process, from the interface to the server-side functionality that powers it. AI has changed the tasks for many software developments services, and full-stack is one of them. The integration of Generative AI (GAI) into full-stack applications presents a transformative opportunity for healthcare assistance by enhancing conversational user interfaces. This study explores the development of an AI-powered healthcare assistant using a React.js frontend, a Node.js/Express.js backend, and Google GAI for symptom analysis and response generation. The aim is to assess the feasibility, performance, and effectiveness of Generative AI in providing real-time medical suggestions. A process of repeating development occurs and the system continuously improves through testing alongside user engagement. Users can maintain safe communication between systems through Jason Web Token (JWT) authentication and HyperText Transfer Protocol Secure (HTTPS) encryption features in addition to multi-language compatibility. Test results show that symptoms processed by the AI-enabled chatbot require 1.5 to 2.5 seconds before providing recommendations with 85-90% accuracy which surpasses traditional rule-based chatbots. The obtained results indicate how GAI shows promise to expand healthcare availability to wider populations.
Authors - Aneri Pandya, Killol Pandya, Hemant Yadav Abstract - Convolutional Neural Networks (CNNs) play a crucial role in medical image analysis, particularly in the detection of pneumonia, by facilitating accurate diagnosis and treatment planning. This paper presents a robust approach to pneumonia classification using the publicly available RSNA Pneumonia Detection Challenge dataset. The proposed method employs an ensemble-based CNN framework that integrates multiple pre-trained architectures, specifically MobileNetV2, InceptionV3, and ResNet152V2, through a combination of stacking techniques and soft voting. This ensemble strategy aims to leverage the individual strengths of each model for improved performance. All models are trained and evaluated exclusively on the RSNA dataset, and their effectiveness is assessed using AUC, accuracy, and F1-score metrics. The results demonstrate that the ensemble consistently outperforms individual models in terms of prediction stability and classification accuracy. This work highlights the potential of deep model aggregation techniques to enhance diagnostic reliability in clinical pneumonia screening systems.
Authors - Sana S Santhosh, Ashitha Mohan, Vishnu N Abstract - Artificial Intelligence (AI) has transformed the cosmetics market by providing tailored suggestions based on client interests and behaviors. This study investigates the correlation between AI-driven personalization and consumer trust, emphasizing critical elements such as the precision of AI recommendations, transparency, and the degree of customization. Research indicates that AI-driven customisation markedly improves client trust, therefore influencing consumer pleasure and brand loyalty. Trust development in AI-driven recommendations is contingent upon perceived reliability, security, and transparency. The review emphasizes that AI transparency techniques, including explainable AI (XAI), enhance customer confidence, whereas excessive personalization may provoke privacy issues. Moreover, the precision of AI recommendations is strongly linked to consumer happiness and intents to repurchase, enhancing perceived expertise and diminishing decision-making anxiety. Brands utilizing AI-generated cosmetic advice must equilibrate customisation, openness, and security to cultivate enduring consumer trust. Future research ought to investigate the ethical ramifications of AI personalization and analyze the evolution of confidence in AI-driven cosmetic brands over time. With the ongoing advancement of AI technology, it is imperative to optimize algorithms for precision and ethical concerns to maintain client confidence in the beauty sector.
Authors - Amarnath K R, Nandini Nayakudi, Lekha S Nair Abstract - Accurate and timely identification of insect species is a foundational requirement for precision agriculture, ecological monitoring, and integrated pest management. This paper presents a novel deep learning architecture for fine-grained acoustic classification of insect species using multichannel wingbeat recordings. Targeting critical agricultural use cases, the model is designed to differentiate between two morphologically and acoustically similar pest species, Halyomorpha halys and Nezara viridula, and a beneficial pollinator, Episyrphus balteatus. The proposed approach utilizes log-Mel and Per-Channel Energy Normalization (PCEN) spectrograms as input features and integrates a residual convolutional backbone with Squeeze and Excite blocks and Transformer inspired self-attention modules to enhance spectro-temporal representation learning. Evaluation on a curated subset of the InsectSound1000 dataset demonstrates a classification accuracy of 88.73% and a macro averaged F1 score of 88.78%. Class wise analysis reveals that the model effectively captures subtle differences in wingbeat harmonics, achieving an AUC of 1.00 and AP of 0.99 for the pollinator class. The results suggest strong discriminative power across ecologically distinct classes and highlight the model’s potential for deployment in automated, non-invasive insect monitoring systems for real-time agricultural decision-making.
Authors - Nathalia Zacarias, Denys Huacanca, Edgar Ramos, Ines Tarazona, Donavan Fuqua Abstract - Supply Chain Planning in agri-food supply chains faces increasing challenges due to global disruptions and resource constraints. Supply Chain 5.0, driven by human-centered approaches, enhances resilience and sustainability by aligning dynamic demand strategies with technological advancements in uncertain environments. To address these complexities, this study applies the Best- Worst Method (BWM) to prioritize key strategic criteria across four dimensions, enabling consistent and structured decision-making. The results emphasize the relevance of advanced supply chain planning and human-centered innovation in fostering adaptive resilience and shaping a more sustainable supply chain. These findings support the transition toward resilient, sustainable, and human-driven agri-food supply chains aligned with the Supply Chain 5.0 framework.
Authors - Windy Permata Suyono, Marsellisa Nindito, Dwi Handarini, Hera Khairunnisa, Eka Septariana Puspa, Surya Anugrah, Rochma Sudiati, Rio Firnanda, Sabo Hermawan, Septi Nurmalita Abstract - The transformation of auditing through Artificial Intelligence (AI) has redefined the way auditors perform their tasks in the digital era. This study aims to examine the role of digital audit and auditor competence in enhancing audit effectiveness, with technological infrastructure as a moderating variable. Using a systematic literature review approach, findings show that both digital audit and auditor competence have a significant positive impact on audit effectiveness. However, technological infrastructure exhibits a dual moderating role. It negatively moderates the relationship between digital audit and audit effectiveness, suggesting that inadequate infrastructure may hinder the benefits of digitalization. Conversely, it positively moderates the relationship between auditor competence and audit effectiveness, indicating that skilled auditors perform better when supported by robust infrastructure. These findings emphasize the need for a holistic strategy that combines advanced technology, auditor upskilling, and infrastructure readiness. This study contributes to the body of knowledge on audit digitalization and supports Sustainable Development Goal 9 (SDG 9) by promoting industry innovation and resilient infrastructure.
Authors - Amita Verma, Simran Sankhyayan, Kritika Sheoran, Rukhsaar Dhaliwal Abstract - Artificial Intelligence has evolved the operation and functioning of every sector and banking is no exception to it. The inclusion of AI has revolutionized banking operations to a significant extent, following the introduction of Automated Teller Machines and Online Banking, transforming traditional banking practices in India. This contribution intends to unravel the complex yet pressing intersections between technology and Banking. This paper examines how AI-powered credit scoring systems, compared to conventional methods, are utilized by banks and financial institutions to accurately assess credit scores and predict credit risks. Credit scoring has been completely transformed by artificial intelligence, which offers sophisticated models and predictive analytics that promise increased efficiency and accuracy. This paper discusses the benefit of leveraging AI technology in credit scoring in terms of financial inclusion of underserved populations of the country paving the way for the establishment of an equitable society. Nevertheless, there are drawbacks and restrictions of using Algorithm technology in credit scoring. This paper highlights the main legal issues and challenges, with particular attention to issues with interpretability, data privacy, data breaches, etc. The principles of fairness, accountability, and transparency are invoked, and the possibility of unintentional biases is called into question by this lack of explainability in AI credit scoring tools. All such issues are then weighed in the context of the existing legal framework in India which eventually highlights the grey areas in the legal realm that the concerned authorities like the legislature, RBI, and financial institutions should ponder upon to ensure the balance between innovation and justice.
Authors - Howard Prioleau, Saurav K. Aryal, Jeremy Blackstone Abstract - This study contributes to ongoing efforts in agricultural technology by advancing the application of artificial intelligence in poultry disease classification, crucial for sustaining agriculture amidst global demographic changes. Focusing on the Chicken Disease Image Classification task, the research evaluates and finetunes state-of-the-art Convolutional Neural Networks (CNNs) like EfficientNet and ConvNeXT, alongside Vision Transformers (ViTs) such as DeiT and Swin Transformer. These models, pretrained on ImageNet, are adapted to classify various poultry diseases from chicken fecal matter images, representing conditions like Coccidiosis, Healthy, New Castle Disease, and Salmonella. The study juxtaposes these advanced models with traditional machine learning approaches to assess their relative efficacy in handling class imbalances and complexities inherent in the dataset. Through a comprehensive evaluation using Precision-Recall curves and Confusion Matrices, the research sheds light on the nuances of model performance in disease identification. The findings underscore the challenges posed by imbalanced datasets and emphasize the need for balanced data in training machine learning models for disease classification. This work showcases the capabilities of current state of the art AI in agricultural settings, aiming to enhance early disease detection tools for poultry health management, especially beneficial in resource-limited environments.
Authors - Sarvani Kocherlakota, Poojitha Panchakarla, Padmanabhuni Venkata Sai Rushikesh Vardhan, Vecha Guhana Naga Gayathri Abstract - E-commerce and supply chain management has emerged as a prominent research topic in recent decades, with numerous findings published in leading international journals. To investigate its current research landscape and emerging trends, we conduct a Bibliometric analysis by using R software 4.5.0, and VOSviewer to identify the research trend, emerging topics, and collaborations among countries, authors, documents and the contributing academic journals during the study period. Sustainability (Switzerland) journal is having the highest publications and Journal of Operations Management is having the most cited document. Zhang X is the most contributing author and Sun Y is the most cited author. China is the most contributing country with around 45% of total publications and also has high international collaborations. By research gap analysis, the future direction in the domain of supply chain management can be integration of live-streaming, emerging technologies, disruption risks, and consumer behavioral preferences.
Authors - Jaskaran Singh Saini, Dinesh Kumar Saini, Bhawna Chahar Abstract - Smart non-motorized transportation is crucial for the sustainable development of urban areas. An astronomical rise in the number of motorized vehicles, particularly in metropolitan cities, has compromised air quality and traffic conditions. NMT, as an effective mode of transport, is often neglected due to a lack of understanding of the long-term negative consequences of internal combustion engines (ICE) by policymakers and transport planners. The rising fuel costs and demand have inflated India's import bills and are expected to surge further until corrective measures are taken. NMT is a low-hanging fruit that can be implemented in urban hotspots at a relatively lower capital expenditure. It can help reduce the demand for fossil fuels and save the state enormous sums that would otherwise have been spent on capital-intensive carbon sequestration technologies. This study aims to highlight the significant economic benefits of NMT, based on data collected from 599 respondents in New Delhi and Chandigarh. The analysis reveals that, in addition to establishing the complex infrastructure of NMT, transport planners and policymakers must foster attitudinal change among citizens, i.e., the soft infrastructure.
Authors - Adithi Varma PC, Venugopal J S, Mayadevi Thampatty PC Abstract - This paper explores the intricate relationship between financial literacy and health insurance as key determinants in fostering financial inclusion among middle-aged class. By examining the mutual connections and co-influence between these concepts, the study aims to uncover how enhancing the health literacy to facilitate greater financial inclusion. Financial Literacy empowers individuals to make sound decisions regarding health insurance, risk management, healthcare cost, and insurance coverage. Conversely, Adequate health insurance which suits your health plan act as financial safety net, reducing huge health-related expenditures and promote financial stability. This study is conducted using a descriptive and analytical design, surveyed 101 respondents in Kozhikode District via convenient sampling through interview schedule. Data analysis involved mean scores, Chi-square, and t-tests. This research proposes that by integrating financial education with health literacy programs we can a create a synergistic effect that leading to an improved health outcome and enhanced financial well-being. This study will facilitate the development of comprehensive strategies that address the financial and health challenges faced by individuals and thereby promoting financial inclusion.
Authors - Selvamani K, Kanimozhi S, Muthusundar SK, Saranraj G, Jagadeesh K Abstract - With the advancement of quantum computing, conventional cryptographic approaches are losing strength against possible attacks, showing the necessity of secure alternatives such as lattice-based cryptography, especially for block chain and IoT implementations. Although lattice-based cryptography provides strong quantum resistance, its practical application is hindered by high computational overhead, intricate key management, and power consumption in resource-constrained IoT devices. Additionally, using it in block chain demands secure signature schemes and authentication protocols without compromising performance. Existing solutions compromise between security and efficiency at the expense of power consumption and higher latency. In an effort to solve these challenges, we present a more robust latticebased cryptographic scheme that enhances aggregate signatures for low communication overhead without compromising security. Our solution also provides a light-weight key exchange mechanism specific to IoT devices, enhancing scalability and energy consumption. Adaptive security elements are also included to counter potential post-quantum threats. Performance tests indicate that our framework substantially minimizes computational complexity while improving security compared to currently available alternatives. The findings attest to its capability to secure block chain-based IoT networks while providing efficient cryptographic operations. By overcoming the major shortcomings of lattice-based cryptography, our research provides a scalable, quantum-resistant solution for IoT and block chain application security in the post-quantum age.
Authors - Selvamani K, Kanimozhi S, Muthusundar SK, Saranraj S, Raju S Abstract - In the contemporary digital environment, organizations across all industries are increasingly dependent on online services to enhance convenience, streamline operations, and boost overall efficiency. While this digital transformation offers undeniable benefits, it simultaneously exposes users and institutions to significant cyber security threats. Malicious actors constantly seek out vulnerabilities in these systems, employing sophisticated methods to gain unauthorized access to sensitive data. One of the most alarming and effective tactics used by cybercriminals is social engineering. Rather than targeting technological flaws, social engineering exploits human behavior, manipulating individuals through deception, coercion, or the illusion of trust to bypass even the most advanced technical security controls. As traditional cyber security defenses such as multi-layered encryption, firewalls, and vulnerability patches grow more robust, attackers are shifting their focus to the human element, which is often the weakest link in the security chain. Unlike technical attacks that require in-depth knowledge of systems and code, social engineering preys on emotions like fear, urgency, curiosity, or helpfulness. These psychological manipulations make it easier for attackers to extract confidential data or financial information without triggering alarms. This paper explores the most common forms of Social Engineering (SE) attacks, including phishing, pretexting, baiting, and tailgating. It also delves into the psychological principles behind these tactics and outlines essential preventative strategies to mitigate the growing threat of social engineering in today’s cyber landscape.
Authors - Iroshan Aberathne, Don Kulasiri, Sandhya Samarasinghe Abstract - Manual techniques are not yet capable of tracking such tiny abnormalities in the early stages of AD due to the inherent complex nature of the structural changes in the brain. Therefore, we propose a novel model agnostic unsupervised feature selection algorithm called Prominent Feature Selection (PFS) to identify significant or vulnerable brain regions via volumetric measurements extracted from MRI images. Features selected by PFS are evaluated in two aspects. First, we compare the results with several other feature selection methods to assess the interpretability of the selected features against age and gender. The experimental results show that the PFS results better describe the structural changes in brain regions against age and gender. Secondly, binary class classification is performed to observe the discriminative power of the selected features against Cognitive Normal (CN) vs. Early Mild Cognitive Impairment (EMCI) subjects. Binary class classification performance measures demonstrate the significance of PFS. It has higher F1 score values despite a lower number of features compared to other feature selection methods. Since, PFS can identify and rank the most significant brain regions at early the stage of AD, diagnosis may become more specific. PFS can also be used as an effective feature selection technique not only to enhance the model performance but also to extend the model’s interpretability which would improve personalized evaluation in areas like eXplainable Artificial Intelligence (XAI).
Authors - Mohammad Nasar, Mohammad Abu Kausar, Md. Abu Nayyer Abstract - Smart farming technologies, particularly image-based crop disease detection systems, are significantly transforming agriculture by enabling farmers to respond promptly to emerging threats. However, deploying centralized AI models in remote or rural farming communities remains challenging due to limited internet connectivity, heterogeneous data sources, and privacy concerns. This paper presents a federated learning (FL) framework designed to support secure and privacy-preserving crop disease detection in decentralized agricultural environments. In the proposed approach, lightweight convolutional neural networks (CNNs) are trained locally on crop images captured by drones or field sensors, ensuring that sensitive data remains at the source. Instead of transmit-ting raw data, only model updates are shared with a central server, which aggregates them using the Federated Averaging (FedAvg) algorithm. To enhance security and model integrity, a cosine similarity-based filter is incorporated to detect and discard malicious or corrupted updates. Additionally, an attention-based heatmap module is introduced to visually highlight disease-affected regions in the crop images, improving system transparency and building user trust. Experimental evaluations on the PlantVillage dataset and real-world drone imagery of wheat and tomato crops demonstrate that the framework achieves an average 4.8% improvement in F1-score under non-IID data conditions com-pared to centralized models, alongside a 9.3% reduction in communication overhead at 50 nodes. This solution is well-suited for resource-constrained rural settings and contributes to scalable, intelligent, and sustainable agricultural digitization.
Authors - Makrand Kadam, Prashant Dhotre Abstract - Active Directory (AD), created by Microsoft, plays a central role in how most organizations manage user identities, access, and system permissions. Because of its importance in controlling access across enterprise networks, it’s a prime target for cyber attackers. This paper takes a closer look at how AD is structured, highlighting key vulnerabilities and common methods used by attackers—like stealing credentials, escalating privileges, using Pass-the-Hash, or creating Golden Tickets. It also examines real-life security breaches to show just how critical AD security has become. Beyond identifying threats, the paper outlines practical defenses such as improved auditing, network segmentation, hardening of domain controllers, and adopting zero trust models. By bringing together current research and best practices from the field, this study offers cybersecurity professionals and system administrators a clear, actionable guide to better protect their AD setups against today’s increasingly sophisticated threats.
Authors - Rajeshwari S B, Sajini G, Jagadish S Kallimani Abstract - In today's fast-paced world, people frequently experience high levels of emotional stress and struggle to access personalized mental health support. Traditional mental health services often face challenges such as limited accessibility and a generalized approach, which can overlook individual emotional nuances. This one-size-fits-all method may fail to address the specific needs of individuals. The need for more responsive and individualized mental health resources has never been greater, highlighting the importance of innovative solutions that offer contextually relevant and personalized care. The Srimad Bhagavad Gita, a timeless spiritual text, offers profound teachings on managing emotions, finding purpose and maintaining balance in life. This project aims to develop a Mental Health Conversational Assistant that provides users with motivational responses rooted in the wisdom of the Bhagavad Gita. This chatbot lever-ages NLP and NLU to detect both the emotional state and the intent behind user queries. It uses advanced techniques like emotion detection and semantic understanding to analyze user input and respond with relevant verses from the Bhagavad Gita. By integrating Hugging Face's emotion detection models and sentence-transformer embeddings, chatbot is capable of providing tailored, contextually relevant verses from Bhagavad Gita that match both the user’s emotions and the meaning of their queries.
Authors - Reena S. Satpute, Nura Muhammad Sani Abstract - Effective disease surveillance and outbreak response are high public health priorities, particularly in resource-limited healthcare systems facing significant challenges with data collection, predictive modeling, and coordinated intervention strategies. It hence proposes an integrated AI-driven framework responding to these challenges and strengthening the capacity of public health authorities to better manage infectious disease outbreaks. This framework will integrate real-time data from different sources such as electronic health records, syndromic surveillance, environmental indicators, and social media in a harmonious way into one centralized repository for the enablement of advanced predictive modeling techniques. It mainly comprises a disease forecasting model, an outbreak detection algorithm, and an intervention optimization tool integrated to provide early warning, accurate prediction, and data-driven decision support. An extensive performance evaluation of the framework was conducted using accuracy, sensitivity, specificity, and area under the ROC curve as some of the metrics. The results are very satisfactory, while the overall accuracy is 94.8% (95% CI: 93.2%-96.4%) with an AUC-ROC of 0.967 (95% CI: 0.951 - 0.983). Comparative performance analysis using state-of-the-art deep learning architectures further establishes the superiority of the proposed hybrid model. Transparency and explainability in the framework were guaranteed up to an extent that allows understanding the models and decision-making process by the public health authorities.
Authors - Adulfo R. Arevalo, Ronald L. Pancho Abstract - This study examines the forecasting performance of three prominent models— Autoregressive Integrated Moving Average (ARIMA), Support Vector Regression (SVR), and Random Forest (RF)—in predicting the short-term price movements of Bitcoin (BTC) and XRP using real daily closing price data from January to June 2024. The research aims to determine which model delivers the most accurate forecasts by applying each technique to actual market data and evaluating their predictive outputs using three standard accuracy metrics: Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Root Mean Squared Error (RMSE). Results show that ARIMA consistently outperformed both SVR and Random Forest in forecasting accuracy, recording the lowest average errors across all metrics for BTC and XRP. While SVR performed poorly in capturing XRP's volatility, Random Forest showed moderate results but lagged behind ARIMA. These findings highlight ARIMA's robustness and practicality as a forecasting tool for short-term cryptocurrency price movements. The study concludes with model-specific recommendations, encouraging the use of ARIMA for near-term price prediction and suggesting further exploration into hybrid and deep learning models for future research.
Authors - Rajeshwari S B, Jagadish S Kallimani Abstract - In the realm of human-computer interaction, the development of input devices accommodating diverse user needs is paramount. Traditional peripherals like mice and keyboards, though effective in typical settings, often falter in specialized conditions such as low-light environments. Addressing these constraints, this study introduces a novel gesture and voice-controlled mouse system. By leveraging real-time video and audio inputs, the system interprets user gestures and voice commands, eliminating reliance on conventional input tools. Additionally, it tackles unresolved interface design issues, striving to optimize user experience and broaden computing resource accessibility. Key to its functionality is the integration of MediaPipe for precise gesture recognition and natural language processing for voice command interpretation. Utilizing standard webcam and microphone hardware enables hands-free cursor control and command execution, particularly aiding users with mobility limitations. Methodologically, the system blends foundational theoretical principles with innovative design concepts for practical realization. Rigorous testing demonstrates a 90% accuracy rate in gesture recognition and a notable 60% accuracy in low-light conditions. Voice command interpretation achieves an impressive 95% accuracy across various accents and speech patterns, with a significant reduction in task completion time com-pared to conventional setups. This research underscores the potential of an affordable and adaptable technology, promising enhanced digital accessibility through ubiquitous webcam and microphone integration.
Authors - Thanh D. Nguyen, Anh T. N. Nguyen Abstract - This study examines how habitual behavior and users’ perceptions of security influence their intention to adopt and use e-wallet services. Building upon the original Technology Acceptance Model (TAM), the research integrates elements from both the Unified Theory of Acceptance and Use of Technology (UTAUT) and its extended version, UTAUT2. Data were collected from 265 valid respondents via an online survey and analyzed using reliability testing, exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and structural equation modeling (SEM). The findings reveal that perceived security, general privacy concerns, and habit all positively affect the intention to adopt e-wallets. Further-more, this intention significantly influences actual usage behavior.
Authors - Nureni Ayofe Azeez, Solomon Akinboro, Isiekwene Chinyere Chioma, Charles Van der Vyver, Awoseemo Babajide Abstract - The propagation of fake news in the digital age has raised momentous concerns about the integrity of information dissemination. This research aims to carry out a comparative assessment of machine learning algorithms for identifying fake news in social media. To achieve this, an attempt was made to collect a wide-ranging dataset of news articles, encompassing both real and fake sources. In the initial phase of the research, we preprocess the news data, extracting relevant features such as text content, metadata, and source credibility. Leveraging a wealth of labelled data, these models learn to differentiate between genuine and fake news articles, while also providing insights into the characteristics that contribute to their classification. In a world where information authenticity is paramount, the implementation of this system provides a valuable tool for discerning consumers of news and enables the swift identification of misinformation sources. The models were trained using four different datasets. Several machine learning algorithms were used for carrying out comparative assessment in terms of their efficiency and accuracy to detect fake news. Selected metrics were chosen for a comparative assessment. At the end of the experiments, for Dataset 1, the Passive Aggressive Algorithm had the highest accuracy of 0.8333, while for Dataset 2, the Passive Aggressive Algorithm had the highest accuracy of 0.914285. However, for Dataset 3, the AdaBoostClassifier Algorithm had the highest accuracy of 0.9142857. Furthermore, in Dataset 4, several algorithms which including Logistic Regression, Passive Aggressive, Ridge Classifier, SGD Classifier, AdaBoost Classifier, Bagging Classifier, Extra Tree Classifier, Gradient Classifier, Random Forest, NUSVC, Linear SVC, Decision Tree, BernoulliNB, and GaussianNB, all had the highest accuracy of 1.0000000. A bar chart was used to represent the results graphically, showing the minimum, maximum, and quartile range.
Authors - Akash Karale, Pratvina Talele Abstract - With the rapid advancement of deepfake audio technology, producing artificial voices that sound remarkably realistic is now simpler than ever. As this becomes available In addition to opening up new creative possibilities, it also poses grave security, disinformation, and fraud concerns. Although many of the deepfake audio detection tools available today are accurate, their application in real-time situations is limited by their high processing overhead. In this work, we introduce a Convolutional Neural Network (CNN)-based lightweight deepfake audio detection model. Our model can detect AI-generated speech accurately and with minimal processing demands by utilizing Mel-Spectrogram features and an effective CNN architecture. The model works well with high accuracy and low latency, according to tests on standard datasets, which makes it appropriate for real-world uses like media verification and voice authentication. Looking ahead, we aim to improve the model's ability to withstand adversarial attacks and adapt to new types of deepfake generation techniques. In the future, we would like to improve the model's robustness against hostile attacks and make it more flexible to accommodate new deepfake generation techniques
Authors - Ram Joshi, Rutuja Khedkar, Prema Sahane, Shantanu Totewad, Nikhil Tarate, Prajodeep Patil, Kedarnath Zunjare Abstract - Deepfakes are manipulated or AI-generated images and videos that appear authentic but are not. A decade ago, detecting deepfakes was relatively straightforward due to the limitations of the technology. However, recent advancements in AI, machine learning, and deep learning have made deepfakes nearly undetectable. Although this technology is often used for entertainment, it is increasingly being used for malicious purposes. The widespread availability of AI tools has further facilitated the creation of convincing deepfakes by fraudsters. Given that people tend to trust what they see and hear, they are particularly susceptible to the influence of realistic deepfakes.
