Authors - John Jenq Abstract - An Artificial Intelligence (AI) agent is an expert that can handle tasks efficiently and accurately, and can execute specific tasks automatically. In this paper, we develop an AI class scheduling assistant system. This system consists of a large language model (LLM) which serves as an interface to the user, and an agent. This agent contains several sub-agents to perform the scheduling task. The data used by the agent was acquired from the University website. It was pre-processed and stored on an excel file that is accessible to the agent. One of the sub-agents will load the file and convert it into Python Pandas data frame. Another sub-agent has the function of finding all the available class sections based on the user’s input from the data frame. This available section information can be used later, along with the user’s data by other sub-agents. OpenAI GPT-4 model is used for the LLM. We implemented the system using Python programming language. According to our experimental results, the agent workflow runs smoothly and quickly, and the system performs the scheduling task accurately.
Authors - Anh Truong, An Nguyen, Huy Le, Thang Pham Abstract - Large language models (LLMs) are being increasingly deployed in customer-service chatbot applications. While traditional fine-tuning approaches often suffer from hallucinations or require computationally expensive retraining, emerging retrieval-based and parameter-efficient methods are increasingly regarded as promising alternatives. However, comprehensive evaluation of these paradigms in a customer-service context remains limited. To address this gap, we conduct a comprehensive comparison of three fine-tuning paradigms - Retrieval-Augmented Generation (RAG), Retrieval-Augmented Fine-Tuning (RAFT), and Weight-Decomposed Low-Rank Adaptation (DoRA) - for customer-service chatbot applications. All methods share a common backbone and are evaluated on the Bitext customer-support dataset. RAG method achieves strong factual consistency at the cost of higher inference latency. RAFT delivers the best overall balance of intent-classification accuracy, coverage, and low hallucination with moderate latency overhead. DoRA extends LoRA (Low Rank Adaptation) by decomposing weight updates into magnitude and directional components for precise low-rank adaptation. On Bitext, DoRA’s end-to-end fine-tuning not only achieves medium accuracy and coverage but also suffers a higher hallucination rate due to limited training data. In larger-scale settings (e.g., SQuAD + 400K-row FAQ), DoRA demonstrates surprising low hallucination and high BLEU/ROUGE, proving that dataset size significantly influences its performance. The results highlight trade-offs between latency, accuracy, and factual reliability, while emphasizing the importance of data scale and retrieval grounding in deploying scalable, trustworthy LLM-based customer-service systems.
Authors - Tahfizul Hasan Zihan, Hana Sultan Chowdhury, Shirazim Munir Deap, Rubayed Mehedi, Farhad Alam, A. M. Shahabuddin, Mahady Hasan Abstract - Recirculating Aquaculture Systems (RAS) have increasingly gained attention in Bangladesh due to their efficient water use. However, maintaining real-time water quality is challenging, as delayed anomaly detection often causes economic loss and compromises fish health. To address this, we developed a low-cost IoT-based monitoring and alert system that integrates industry-standard sensors to track key water parameters (pH, turbidity, dissolved oxygen, temperature, oxidation-reduction potential, total dissolved solids, and electrical conductivity) using Arduino microcontrollers and ESP8266 Wi-Fi modules. The total hardware cost was under USD 700, an order of magnitude cheaper than commercial systems. Sensor readings were published every 15 seconds via MQTT to the Thing Speak cloud platform, enabling real-time visualization and WhatsApp-based alerts for immediate intervention. Experimental deployment demonstrated the system’s capability to maintain optimal conditions consistently, identifying critical events such as abnormal water quality drops efficiently. Clear daily and monthly trends were observed, enabling predictive adjustments and informed automated decision-making. This IoT approach significantly improves operational efficiency and reliability in resource-limited aquaculture settings typical of Bangladesh, ensuring sustained fish productivity and welfare.
Authors - Ayman Alarabiat, Yousef Alarabiat, Mahmoud AlZuabi, Mamoun Shakatreh Abstract - Governments are increasingly leveraging artificial intelligence (AI) Chatbots to enhance e-service accessibility. However, Chatbots adoption among citizens’ remains low, limiting their intended benefits. This study explores citizens’ perspectives on government Chatbots adoption. A quantitative correlational research approach was employed, collecting 358 responses from Jordani-an citizens who had used government Chatbots in the last 6 to 12 months. An online survey measured ten key constructs: low complexity, relative advantage, compatibility, trialability, observability, trust, responsiveness, perceived intelligence, anthropomorphism, and Chatbot adoption. Data analysis using SPSS 24 revealed low willingness to adopt Chatbots, primarily due to concerns about Chatbot intelligence, responsiveness, and trust. Additionally, low observability and limited perceived relative advantage further hinder adoption. These findings provide insights for policymakers, government agencies, and Chatbot developers to enhance Chatbot functionality and user experience. Key recommendations include improving Chatbot intelligence and responsiveness, increasing public awareness, and fostering greater trust in the Chatbot. Addressing these factors can drive greater adoption, maximizing the efficiency and impact of AI-driven public services.
Authors - Maisyaroh, Agus Timan, Mustiningsih, Maulana Amirul Adha, Indra Lesmana, Rudy Ansar, Novia Putri Arianti Abstract - Digital learning transformation in rural schools faces various challenges, including limited infrastructure and low digital literacy among teachers. This study aims to explore the contribution of collegial supervision in supporting this transformation. Using a descriptive qualitative approach, data were collected through in-depth interviews, observations, and documentation at a high-performing private secondary school in a rural area. The findings reveal that collegial supervision is grounded in values of trust, equality, shared reflection, and communal cooperation. This approach provides a collaborative space for teachers to exchange best practices, enhance their technological competencies, and develop effective digital teaching strategies. Despite challenges related to time, resources, and varying levels of understanding, collegial supervision has proven effective in fostering an innovative school culture that supports digital-era teaching and learning. This study contributes to the literature on educational supervision by offering context-specific insights and practical strategies for teacher development in resource-constrained schools.
Authors - Nishat Shaikh, Parth Shah, Bimal Patel Abstract - Deep learning has revolutionized oncology by enabling unprecedent-ed integration of multimodal data for cancer diagnosis and prognosis. This comprehensive survey presents the first systematic analysis of deep learning architectures across four major cancer types (lung, breast, skin, and brain) through three critical data modalities: medical imaging, histopathology, and genomics. Our unique contribution lies in providing a structured taxonomy of multimodal fusion strategies and identifying critical architectural innovations that have emerged in the 2021-2025 period. We systematically analyze 60+ recent studies, revealing that attention-based mechanisms and Transformer architectures demonstrate superior performance in handling heterogeneous cancer data compared to traditional CNN approach-es. Our analysis uncovers three key research gaps: (1) limited interpretability frameworks for clinical deployment, (2) insufficient standardization across institutions, and (3) scalability challenges for real-world implementation. This survey uniquely bridges the gap between theoretical deep learning advances and practical oncological applications by proposing a unified framework for multimodal cancer analysis. We provide actionable insights for researchers and clinicians, establishing clear directions for future development in AI-driven cancer care that addresses both technical innovation and clinical translation requirements.
Authors - Esthefano Palomino, Rodrigo Condor, Edgar Ramos, Ludwig Tocto, Victor Pimentel Abstract - Agri-food supply chains are increasingly exposed to climate variability, resource constraints, and disruptive shocks, requiring strategies that enhance resilience and sustainability. This study applies the Best–Worst Method (BWM) to assess and prioritize four strategic dimensions: resilience, smart manufacturing, innovation, and circularity. The results indicate that resilience is the most critical enabler, while smart manufacturing and innovation provide significant complementary support, with circularity ranking lowest. These findings emphasize the central role of resilience in safeguarding agri-food systems and demonstrate how technology and innovation can strengthen long-term sustainability. The study contributes by offering a structured decision-making framework that helps managers and policymakers focus on the most impactful dimensions when designing more adaptive and sustainable agri-food supply chains.
Authors - Sarika Pabalkar(Wagh), Kirti Jain Abstract - Generative adversarial networks, or GANs, are strong tools for improving medical pictures because they can make high-quality virtual images that can be used to solve problems like limited datasets, image unpredictability, and poor diagnosis accuracy. This research looks at all the ways that GAN can be used in medical imaging and highlights the most important improvements in the process of segmentation reconstruction, disease identification, and cross-modal synthesis. A structured methodology based on PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) was employed to identify, screen, and analyze 42 peer-reviewed articles published between 2020 and 2025 across databases such as PubMed, Scopus, IEEE, and ScienceDirect. The review compares widely used GAN architectures—including CycleGAN, Pix2Pix, DCGAN, and ProGAN— evaluating their strengths, limitations, and suitability for different imaging modalities such as MRI, CT, X-ray, ultrasound, and mammography. There are significant problems, including mode collapse, instability in training, no consistent evaluation standard, and few opportunities for clinical validation. The results show the increasing importance of GANs in generating clinically useful data for rare disease information and for cross-modality tasks. The discussion also indicated future interventions that would focus on model stability, ethical use, clinical incorporation, and generalization across populations. The aim of the review is to help fellow researchers and practitioners through assessing state-of-the-art GAN approaches and defining gaps that will need to be addressed prior to more adaptation of GANs in medical imaging.
Authors - Om Roy, Dhruv Shingala, Priyanka Patel Abstract - Bias within Artificial Intelligence (AI) systems constitutes a profound challenge with substantial implications for fairness, accountability, and societal equity. This paper presents an exhaustive examination of the ontology of bias in AI, delving deeply into its conceptual underpinnings and exploring the intricate algorithmic consequences arising from biased data and models. By establishing a comprehensive and nuanced framework that categorizes diverse manifestations of bias and elucidates their origins, this study aims to foster a profound understanding of how bias permeates AI systems. Integrating interdisciplinary perspectives drawn from philosophy, sociology, and computer science, the ontology of bias is meticulously dissected to reveal its multifaceted nature. Furthermore, the paper investigates the profound impacts of these biases on critical decision-making processes and proposes multifaceted strategies for mitigating bias through ethical design, advanced algorithmic techniques, and stringent regulatory frameworks. Through detailed case studies and empirical analysis, this research highlights the inherent complexities in addressing bias and underscores the imperative for collaborative endeavors to cultivate equitable AI technologies.
Authors - Payel Das, Tejaswini Seelam, Rajeswari Annam Abstract - This study investigates the intersection between generative AI, sustainable development and education by conducting a systematic review on the utilization of ChatGPT and other Large Language Models (LLMs) in relation to Sustainable Development Goals (SDGs), focusing on SDG 4 and SDG 12. The article explores how technology-enhanced learning and LLMs are challenging pedagogical traditions, practices, even ethical dilemmas, and calls for responsible governance. There were 59 peer-reviewed articles included between 2023-2025 which provided data synthesis. Three core clusters were discerned using the VOS viewer keyword co-occurrence map: (1) educational impact and learning outcomes, (2) stakeholder ethics and governance perspectives, and (3) integrity Challenges of AI-assisted instruction. The study also raises alarms about threats that could emerge from misinformation, academic cheating, algorithmic bias and unequal ac-cess. It also underscores the lesser-known environmental footprint of AI tools.
Authors - Mary Anne Q. Zheng, Gerry Paul C. Genove Abstract - This paper presents a global benchmarking analysis to evaluate the Philippines’ infrastructure adaptability to 5G, IoT, and AI. Based on a systematic literature review of 35 peer-reviewed studies, industry analyses, and government reports from 2018 to 2025, it evaluates the current state of deployment, identifies key barriers, and proposes actionable strategies for alignment with international best practices. Results indicate that 5G coverage is expanding in metropolitan areas, IoT applications are emerging in sectors such as utilities and agriculture, and AI integration is concentrated in enterprise and pilot projects. However, socioeconomic inequality, an archipelagic geography, bureaucratic inefficiencies, and fragmented governance emerge as primary constraints on broader adaptability and inclusivity. Comparative analysis against global leaders and selected ASEAN peers, including Japan, South Korea, the United States, Germany, Malaysia, Vietnam, Thailand, and Singapore, highlights the importance of regulatory coherence, coordinated infrastructure investment, and integration with socioeconomic strategies. Policy lessons from peer nations demonstrate how centralized spectrum management, regulatory sandboxes, rural inclusion programs, and digital upskilling initiatives can accelerate adaptation. The findings highlight the need for coherent policies, shared infrastructure models, and capacity-building frameworks to align the Philippines with international best practices.
Authors - Monica Pilco, Lisset Ugaz, Edgar Ramos, Donovan Fuqua Abstract - The volatility and complexity of the marketplace require planning approaches that balance efficiency and adaptability in engineer-to-order (ETO) manufacturing environments. This study brings together lean, agile (Leagile) principles and smart manufacturing planning to improve responsiveness without compromising operational performance. Using the Fuzzy Analytic Hierarchy Process (F-AHP) methodology, this study combines a comprehensive literature review with the evaluation of strategies and drivers in a manufacturing case study. The results highlight the importance of Digital Integration, Digital Value Stream, and Integrated Planning to achieve leagile and sustainable operations. Conclusively, this study offers practical suggestions for leveraging digital drivers to strengthen resilience and long-term sustainable smart manufacturing.
Authors - Wahyu Wastuti Abstract - This study conducts a Systematic Literature Review (SLR) of 30 Scopus-indexed articles published between 2016 and 2025 to examine the conceptualization and development of Islamic Social Enterprise (ISE) and Islamic Social Entrepreneurship (ISEn). The review reveals that ISE is predominantly framed as an organizational form integrating Islamic principles into governance, accountability, and social welfare structures, while ISEn emphasizes individual-driven innovation and entrepreneurial behavior aligned with sharī‘ah values. Despite their differences, both models converge in their pursuit of financial sustainability, social justice, and community empowerment. Thematic analysis identifies four recurring domains in the literature: governance, innovation, sustainability, and maqāṣid al-sharī‘ah. Governance studies highlight ac-countability, transparency, and stakeholder trust, while research on ISEn emphasizes digitalization, creative financing, and entrepreneurial intention. Sustainability is viewed not only in terms of economic viability but also as encompassing ethical and social commitments, with maqāṣid al-sharī‘ah providing a normative framework that guides decision-making and ensures alignment with faith-based objectives. The findings reveal that the conceptual boundaries be-tween ISE and ISEn remain blurred but complementary, suggesting their integration offers both institutional stability and entrepreneurial dynamism. This study contributes to the academic discourse by clarifying definitions, synthesizing thematic trends, and identifying research gaps. Practically, the results offer implications for policymakers in designing supportive ecosystems, for practitioners in building hybrid models, and for academics in advancing theoretical clarity.
Authors - Jaimin Dave, Chintan Shah, Premal Patel Abstract - Intrusion Detection Systems (IDS) have been instrumental in preventing emerging threats from compromising a network. One of the key challenges in the field is to design an IDS with high detection coverage and low false positives. This study investigates the potential of hybrid deep learning systems to address this challenge. We developed a CNN+LSTM model that leverages the spatial feature extraction power of CNN and the temporal sequence learning capacity of LSTM. Experimental results demonstrate that the proposed CNN-LSTM model consistently outperforms LightGBM and XGBoost, achieving the highest accuracy (92.7%), precision (89.5%), recall (81.8%), and F1-score (84.3%), thereby confirming its robustness and effectiveness for intrusion detection. Additionally, comprehensive assessments indicate that for more complex and hidden attack patterns, the CNN+LSTM model is more advanced in terms of detection availability and effectiveness. The relevance of this work emphasises the need to modernise IDS systems by using hybrid deep learning techniques. Furthermore, the combination of convolutional neural networks with LSTM improves detection capacity and supports feature-level detail for new hazards.
Authors - Lukas Herout, Arvind Panwar Abstract - This paper explores the adoption of Zero Trust Architecture (ZTA) in small and medium-sized enterprises (SMEs) in the Czech Republic, aiming to identify key barriers, enabling factors, and readiness levels for its implementation. Drawing on survey data from 230 SME respondents, we investigate aware-ness of ZTA principles, current security practices, and the perceived value of specific support measures. The study is guided by clearly defined research questions and hypotheses. We construct a composite readiness index based on technical and organizational measures and analyze its relationship with firm size, security maturity, and training frequency. Key findings indicate that while ZTA awareness is relatively low (only 43.4% had heard of the term), many companies have already implemented foundational elements aligned with ZTA. Security readiness is significantly higher among firms with formal security policies and regular employee training. Medium-sized enterprises show notably greater implementation levels than micro and small firms. The most frequently requested forms of support include practical implementation guides, financial aid, and consultations with experts. These preferences vary by firm size, with smaller firms emphasizing financial support and larger ones favoring professional services. The paper concludes with a roadmap for SME-oriented ZTA adoption, emphasizing the need for scalable, low-barrier solutions and targeted support. Our findings provide empirical insights to inform policy design, training programs, and vendor strategies for secure digital transformation in the SME sector.
Authors - Hasin Mahir, Tahfizul Hasan Zihan, Md. Shirazim Munir, M. M. Kamal, Mahady Hasan, Md. Tarek Habib Abstract - Accurate Indoor Air Quality (IAQ) forecasting can support affordable and privacy-conscious interventions in classrooms and workplaces. However, most existing systems depend on cloud-based inference, which introduces additional cost, network dependency, and reliability issues. In this work, we present a comparative evaluation of three ensemble tree models—Random Forest, XGBoost, and LightGBM—implemented directly on an ESP32-S3 microcontroller for edge-based IAQ prediction of CO2 and PM2.5. A unified feature extraction strategy was employed, combining temporal encodings, lagged variables, and short rolling aggregates. Models were trained offline, converted into C, and embedded into the device firmware. Their performance was assessed with respect to (i) forecasting accuracy, using MAPE and RMSPE, and (ii) resource efficiency, including flash storage, RAM usage, and inference latency. Experimental results indicate that boosted methods (XGBoost and LightGBM) provide higher accuracy than Random Forest while staying within the strict memory and timing limits of the ESP32. This demonstrates the feasibility of fully self-contained IAQ forecasting on ultra-low-cost hardware, without reliance on cloud resources. Furthermore, we outline trade-offs between model complexity and efficiency, and provide a reproducible toolchain for exporting tree ensembles into deployable ESP32 firmware, highlighting their practicality for resource-constrained IAQ monitoring.
Authors - Amogh Kotha Nagaraj, Rajesh Eswarawaka, Aadi B Kumar, Sujay D Abstract - This paper will examine the synthesis of fuzzy logic, neural network applications and hybrid computational techniques with quantum computing for data-driven system improvement. The intersection of these innovations will provide exciting directions for intelligent decision-making, adaptive learning, and effective data processing. Fuzzy logic offers a mechanism for dealing with uncertainty, whereas these hybrid techniques blend the powers of multiple paradigms of computation. Neural networks provide the capability of deep learning, and quantum computing provides new means for optimization and parallelism. In this work, the theoretical background, real-world applications, and prospects of quantum-assisted hybrid intelligence systems are presented. The results imply that combined approaches can greatly enhance performance in high-complexity data environments, opens up opportunities for future intelligent systems.
Authors - Pimal Khanpara, Prasham Shah, Sharada Valiveti, Gaurang Raval Abstract - This article examines recent advancements in log anomaly detection, a key factor in ensuring system reliability and security. Research is categorized into four main approaches like traditional machine learning, deep learning, transformer-based models, and other emerging methods leveraging large language models (LLMs). Evaluations across benchmark datasets such as HDFS, BGL, Thunderbird, and Spirit reveal a steady progression in detection capabilities, with LLM-based approaches generally outperforming earlier methods. Techniques like LogRAG, LLMeLog, and EagerLog achieve F1-scores above 99% on certain datasets, marking substantial improvements over traditional baselines. Current research trends include multimodal analysis, few-shot and zeroshot learning, explainable AI, lightweight models, and deeper semantic understanding of logs. These directions are increasingly important as modern systems generate massive, diverse log volumes that demand more scalable and intelligent anomaly detection solutions. However, persistent challenges like log parsing accuracy, effective feature extraction, and model adaptability to evolving log formats continue to limit performance in real-world applications. This review provides a structured overview of current methods, highlights the strengths of LLM-based techniques, and outlines promising areas for future research to advance the field of log anomaly detection.
Authors - Sudarshan Sharma, Amita Chauhan, Sakshi Kaushal Abstract - Unmanned Aerial Vehicles (UAVs) have revolutionized industries such as defense, agriculture, and logistics due to their adaptability and ease of deployment. However, their growing dependence on wireless communication protocols, including MAVLink, UAVCAN, and UranusLink, introduces significant security challenges. This research offers a comprehensive security evaluation of the MAVLink protocol through experimental simulations in Software-In- The-Loop (SITL) environments. The study simulates Distributed Denial-of-Service (DDoS) attacks at both the Transmission Control Protocol (TCP) and User Datagram Protocol (UDP) layers to identify protocol vulnerabilities. Captured network traffic is compiled into a structured dataset and analyzed using machine learning algorithms, specifically Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbor (KNN) algorithms, to differentiate between normal and malicious packets. The findings highlight critical security weaknesses in MAVLink communications and introduce a robust simulation and classification pipeline designed to enhance the effective detection of attacks, thereby improving UAV security.
Authors - Suparna Ray, Sushruthan Babu, Niranjan Babu Abstract - Our study explores the transformative role of Blockchain Technology (BCT) in capital markets (CM) employing a mixed-method research design. We synthesize extant research and analyse global regulatory case studies, particularly focusing on the Indian context. Utilising both bibliometric analysis and qualitative case study research, our paper highlights the evolution of academic dis-course, publication trends, and practical applications of Distributed Ledger Technology (DLT) across equities, debt, and derivatives markets. Our findings indicate that while BCT promises efficiency, transparency, decentralization, and cost reduction, its large-scale adoption remains constrained by regulatory uncertainty, scalability concerns, and institutional resistance. This paper emphasizes the emerging regulatory frameworks in the US, EU, and Asia-Pacific, acknowledging diverse approaches ranging from innovation-friendly sandboxes to cautious restrictions. We find limited empirical evidence on systemic risk, tokenization, liquidity, and ESG-linked instruments; hence, we propose future research avenues that integrate finance, law, computer science, and public policy. Through coherent mapping of the global regulatory landscape and reviewing scholarly contributions, our paper contributes to understanding how BCT is reshaping CM infrastructure and provides pathways for policymakers, regulators, and market participants toward responsible and sustainable digital transformation.
Authors - D. S. Vangao, Pallavi Khatri Abstract - Image forgeries have become much more common due to the widespread use of sophisticated digital image manipulation tools, which makes it difficult to independently confirm authenticity in domains like social media, digital forensics, and journalism. This survey study paper offers a thorough analysis of the most recent methods for identifying image modifications including copy-move and splicing forgeries, with an emphasis on combining Convolutional Neural Networks (CNNs) with Error Level Analysis (ELA). We evaluate important approaches from recent literature based on their architectures, datasets, and performance metrics. We do so via methodically analyzing both contemporary deep learning techniques and traditional methods that rely on handcrafted features. Our investigation shows that hybrid ELA-CNN techniques routinely beat alternative techniques, attaining accuracy rates exceeding 90% on common datasets such as CASIA. More reliable, real-time solutions are required, nevertheless, as problems like identifying subtle tampering and increasing computational effectiveness continue to arise. We wrap up by discussing the present state of the field's limits and potential avenues for future research in digital image forensics.
Authors - Chethana R.M., S.P. Manikandan Abstract - Organisations around the world face never-before-seen difficulties as a result of the quickly changing cybersecurity landscape, which calls for advanced analytical techniques to recognise and anticipate threat trends. In order to analyse current cyberthreats and their organisational impact, this study uses sophisticated statistical techniques such as ANOVA, regression analysis, machine learning classification, and clustering algorithms. We examine threat distribution patterns, financial impact correlations, and predictive modeling capabilities for ten key threat categories malware, phishing, social engineering, ransomware, insider threats, supply chain attacks, DoS/DDoS attacks, zero-day exploits, credential theft, and AI-powered attacks by thoroughly analysing 360 cyber security incidents that occurred across several Indian cities between 2019 and 2024. The most significant predictor of financial impact, according to our statistical analysis, is incident type (F=4.58, p
Authors - GGS Pradeep, Thrilok. Kolla, Rajesh Sharma R, Akey Sungheetha, U Ananthanagu, Pellakuri Vidyullatha Abstract - This paper outlines a holistic framework for improving query comprehension in information retrieval (IR) systems through deep contextual embeddings. Using Sentence-BERT, we transform the queries and documents into dense semantic representations and then calculate cosine similarity for the relevance measure. We especially emphasize interpretability by adding various statistical and visualization techniques, like heatmaps, KDE plots, UMAP projection, dendrograms, and box plots, to identify and analyze latent semantic patterns and keep close tabs on the inner workings of how different documents relate to such varied query formulations. We also follow an avenue of the interquartile dispersion and distributional behavior of similarity scores to understand the embedding consistency and discriminative power. This methodology, therefore, encompasses both retrieval accuracy and explainability as it provides visible insights into improving query reformulation, semantic search, and content recommendation. Experimental visualizations have further demonstrated their effectiveness towards deep semantic alignment, especially during the processing of complex multi-topic corpora. The abstract is technically solid and easily comprehensible because it is rationally structured and clearly states the research’s methodology, instruments, and objectives.
Authors - GGS Pradeep, Thrilok. Kolla, U Ananthanagu, Akey Sungheetha, Rajesh Sharma R Abstract - Predicting non-stationary time series is still a tough nut to crack: the statistical properties may vary, and so will the temporal dynamics. In this context, a robust and interpretable approach involving the integration of a Takagi–Sugeno neuro-fuzzy inference system and evolutionary multi-objective optimization is proposed. The NSGA-II optimization procedure minimizes prediction accuracy alongside model complexity, evolving the fuzzy rule base and membership parameters and the linear consequents. Contrary to the conventional opaque models, a neuro-fuzzy architecture affords some degree of transparency, letting one see how the functional temporal patterns have been learned. Results of the experiments conducted demonstrate the ability of the system to adapt to structural changes in the data, with a corresponding better capability of reducing residual errors and increasing model compactness using a synthetic non-stationary dataset. Further, residual oscillations and fuzzy rule surfaces, alongside the convergence behavior and the Pareto optimality, present powerful evidence in support of the proposed model’s effectiveness and interpretability. The proposed framework represents a potent alternative for real-time adaptive forecasting in dynamic environments where precision and explainability are both crucial. Simulated non-stationary time series will be good enough to justify the approach of lag-embedding. Thoughtful integration of interpretable fuzzy logic and adaptive learning via evolutionary optimization would strengthen the contribution.
Authors - Ritu Raut, Sudhir Agarmore Abstract - The kick to stride the new era in wireless communication begins once 5G technologies holding the promises of speeds previously unheard of and extremely low latency with the capacity to connect a large number of devices at once [4]. This article reviews the potential of 5G in transforming such industries as healthcare, driverless cars, smart cities, and the Internet of Things. It discusses enabling technologies such as millimeter waves, massive MIMO (Multiple Input Multiple Output), and network slicing [1], among others, and also its infrastructure, security, and the possibility of availability of the frequency spectrum. For this purpose, this research aimed to provide an inclusive overview of how this new generation of wireless technology may impinge upon the future of worldwide communication and connectivity by assessing both the benefits and challenges that emanate from 5G.
Authors - D. S. Vangao, Pallavi Khatri, Diya Khatri Abstract - In the digital age, the proliferation of image manipulation has raised significant concerns regarding the authenticity and integrity of visual content. This study, explores the critical role of image forensic tools in detecting digital forgeries and ensuring the credibility of photographic evidence. A comprehensive evaluation of a range of widely used image forensic tools—including but not limited to FotoForensics, JPEG snoop, Izitru, Forensically, and Amped Authenticate—assessing their performance in identifying various types of tampering, such as splicing, cloning, resampling, and compression artifacts is done in this work. Each tool is evaluated on the basis of criteria such as detection accuracy, supported forgery type, user interface, automation capabilities, and analysis depth. The comparative analysis reveals strengths and limitations unique to each tool, offering insights into their suitability for different investigative scenarios. Our findings highlight that no single tool excels universally; rather, a combination of tools often yields more reliable results. This study underscores the necessity for cybersecurity professionals, digital investigators, and media analysts to adopt a multitool strategy for robust image authentication.
Authors - GGS Pradeep, Thrilok. Kolla, Rajesh Sharma R, Akey Sungheetha, N Vijayalakshmi, Pellakuri Vidyullatha Abstract - The early identification of neurodegenerative diseases, like Alzheimer’s and Parkinson’s disease, is crucial for their management, although difficult due to subtle, heterogeneous, and evolutionarily conditioned clinical patterns. In this paper, we present a fuzzy-neural hybrid that incorporates the beneficial properties of fuzzy inference systems and temporal deep learning toward improved early diagnosis. Our model analyzes longitudinal multimodal medical data (neuroimaging and clinical scores), combining fuzzy rule-based feature encoding with Long Short- Term Memory (LSTM) networks for temporal pattern extraction. Interpretability is provided by assigning Gaussian membership functions, while the LSTM component encodes the dynamics of disease progression. The proposed system is validated on synthetic and benchmark datasets, showing robust classification performance and excellent temporal tracking of patient trajectories. Interpretations of internal representations are provided through visualizations such as confusion matrices, t-SNE embeddings, and 3D PCA trajectories. The results indicate that the hybrid approach provides both powerful predictions and transparent operation, making it an excellent option for clinical decision support in neurodegenerative diagnostics. The authors propose a hybrid diagnostic system that uses Gaussian fuzzy membership functions for interpretable feature encoding and LSTM networks for temporal sequence modeling of multimodal medical data. The authors validated their method with synthetic data as well as benchmark datasets, achieving perfect binary classification in their experimental results.
Authors - Agus Timan, Maisyaroh, Maulana Amirul Adha, Indra Lesmana, Anabelie Villa Valdez, Arum Sri Banowati Abstract - The urgency of climate change, environmental degradation, and the need for sustainable development has placed education at the forefront of global transformation. As schools evolve into eco-conscious institutions, there is growing interest in integrating digital technologies to enhance sustainability-driven governance. This has led to the emergence of Green Digital School Management (GDSM), a model that synergizes environmental responsibility with technology-based school leadership. This study aims to systematically explore the core com-ponents, implementation challenges, and effective strategies of GDSM by con-ducting a Systematic Literature Review (SLR) based on PRISMA guidelines. An initial pool of 892 articles was screened, 87 high-quality studies were selected for thematic analysis. The findings indicate that GDSM encompasses digital environmental policies, IoT-based monitoring, and participatory platforms that engage school communities in green practices. Despite significant barriers, including infrastructure gaps, limited digital-environmental literacy, and policy frag-mentation, several innovative strategies have emerged, such as digital teacher training, cross-sector partnerships, and gamified environmental education tools. This review contributes to the conceptual development of GDSM and provides actionable recommendations for policymakers, educators, and technology providers seeking to build smart, green, and sustainable school ecosystems.
Authors - Tsunenori Inakura, Shotaro Imai, Kunihiko Takamatsu, Sayaka Matsumoto, Masao Mori Abstract - Institutional Research (IR) in universities needs data from many different systems, but it is often hard to know what data is stored and where. To solve this, it is important to design both databases and business processes together. In this study, categorical databases, ontology logs (ologs), and event driven process chains (EPCs) are used to connect institutional documents with data and process design. Ologs describe concepts and their functional relations in a formal but also readable way, while EPCs show how events and functions go step by step in the real processes. Generative AI was used to support both tasks. The AI helps to read documents, to extract concepts and functions, and to check consistency. By comparing ologs and EPCs, both sides can be improved and give a unified view of data and processes. This makes it possible to design information systems for IR with more consistency and less effort.
Authors - Vijay Siva, Vijayakumar Ponnusamy Abstract - According to the United Nations, global population will rise from 7 billion currently to 9 billion in 2050. The world will need far more food, and agriculture will face remarkable pressure to meet demand. Emerging application of deep learning in agriculture includes ripeness detection of tomato which harvest tomato in appropriate time and prevents rotten of tomato. This saves the framer from loss. Tomato was classified as ripe, half-ripe, and unripe, based on maturity stage of fruit. Various deep learning methodologies were analyzed for maturity detection of tomato. Multimodal approach for maturity detection is suggested for enhanced accuracy in ripeness detection.
