Pandit Sundarlal Sharma Open University

Wildfire forecasting is a critical challenge in environmental signal processing and disaster response planning. The ability to interpret multimodal spatiotemporal signals is essential for early warning systems and resource deployment. This study addresses these limitations by proposing a unified prediction-to-action framework. We utilized four open-access datasets—wildland fire emissions database (WFED), fire information for resource management system (FIRMS), Sentinel Hub, and a custom moderate resolution imaging spectroradiometer+shuttle radar topography mission (ERA5+MODIS+SRTM) fusion—covering fire occurrences, vegetation indices, meteorological parameters, and topographic features. These heterogeneous signals were preprocessed, aligned, and transformed into structured tensors for model training and evaluation. We use a transformer-based system to understand long-term patterns in space and time, enhanced by a belief–desire–intention (BDI) reasoning module that connects our predictions to flexible wildfire response plans. The novelty lies in the integration of signal-aware attention mechanisms with symbolic decision modeling. Model performance was evaluated using F1-score, intersection over union (IoU), mean absolute error (MAE), and directional accuracy. The suggested framework did better than the basic convolutional neural network (CNN) models, reaching an F1-score of 0.74, a directional accuracy of 84.3%, and lowering the MAE to 7.6 km², while also providing clear and relevant action suggestions.

Federated Learning (FL) has become a disruptive paradigm for decentralized training of models while still being able to preserve data privacy. Personalized Federated Learning (pFL) [4] extends FL to handle non-i.i.d. user data distributions often observed in practice. However, under extreme non-IID conditions, most pFL methods suffer from reduced performances as global and personalized components are mixed up. In this paper, we try to tackle the problem of statistical heterogeneity by trying to learn a robust learning framework that aims at differentiating personalized aspects from global knowledge during training. We also present results for non-IID partitioned data on Dirichlet distributed FMNIST, CIFAR-10, and CIFAR-100 benchmark datasets. Our proposed approach, HD-pFL, leverages a generative β-VAE model for unsupervised feature representations depersonalization and introduces a dual head architecture to disentangle the global and personalized classifier components. The primary novelty is the subtraction-based depersonalization that filters client-specific biases from shared representations. We measure performance via global accuracy, personalized accuracy, client fairness (measured by the standard deviation), and communication cost. Experimental results demonstrate that HD-pFL consistently outperforms state-of-the-art baselines across all datasets, achieving improved accuracy, faster convergence, enhanced fairness, and communication efficiency in highly heterogeneous federated environments.

In modern sociology, the study of technology has become extremely important for understanding social structure, social behavior, and social change. Technology is not merely a tool or a machine; rather, it is a dynamic social force that influences economic, political, cultural, and family structures. It highlights the reciprocal and co-evolutionary relationship between society and technological innovations, in which the needs and values of society determine the direction of technological development, while technological changes reshape social behavior, institutions, and lifestyles. Technology has accelerated processes such as economic productivity, employment generation, digital education, e-commerce, online communication, and global cultural exchange. At the same time, it has also created challenges such as social inequality, the digital divide, cultural homogenization, and environmental crises. In sociology, technology has opened new possibilities for understanding social trends more accurately through the adoption of modern tools for research methodology, data collection, and analysis. This study makes it clear that establishing a balanced and responsible relationship between society and technology is essential to ensure social development, cultural preservation, economic progress, and environmental sustainability. Abstract in Hindi Language: आधुनिक समाजशास्त्र में प्रौद्योगिकी का अध्ययन समाज की संरचना, सामाजिक व्यवहार और परिवर्तन को समझने के लिए अत्यंत महत्वपूर्ण हो गया है। प्रौद्योगिकी केवल उपकरण या यंत्र नहीं है, बल्कि यह एक गतिशील सामाजिक शक्ति है जो आर्थिक, राजनीतिक, सांस्कृतिक और पारिवारिक संरचनाओं को प्रभावित करती है। यह समाज और तकनीकी नवाचारों के बीच परस्पर और सह-विकास संबंध को उजागर करती है, जिसमें समाज की आवश्यकताएँ और मूल्य तकनीकी विकास की दिशा निर्धारित करते हैं, और तकनीकी परिवर्तन समाज के व्यवहार, संस्थाओं और जीवनशैली को पुनर्गठित करते हैं। प्रौद्योगिकी ने आर्थिक उत्पादकता, रोजगार सृजन, डिजिटल शिक्षा, ई-कॉमर्स, ऑनलाइन संचार और वैश्विक सांस्कृतिक आदान-प्रदान जैसी प्रक्रियाओं को तीव्र गति दी है। इसके साथ ही सामाजिक असमानता, डिजिटल डिवाइड, सांस्कृतिक समरूपता और पर्यावरणीय संकट जैसी चुनौतियाँ भी उत्पन्न हुई हैं। समाजशास्त्र में प्रौद्योगिकी ने अनुसंधान पद्धतियों, डेटा संग्रह और विश्लेषण के आधुनिक उपकरणों को अपनाने के माध्यम से सामाजिक प्रवृत्तियों को अधिक सटीक रूप में समझने की संभावनाएँ प्रदान की हैं। इस अध्ययन से स्पष्ट होता है कि समाज और प्रौद्योगिकी के बीच संतुलित और जिम्मेदाराना संबंध स्थापित करना अति आवश्यक है ताकि सामाजिक विकास, सांस्कृतिक संरक्षण, आर्थिक प्रगति और पर्यावरणीय स्थिरता को सुनिश्चित किया जा सके। Keywords: समाजशास्त्र, प्रौद्योगिकी, सामाजिक परिवर्तन, डिजिटल डिवाइड, सांस्कृतिक परिवर्तन, आर्थिक विकास, और सामाजिक संस्थाएँ

The object here for IoT in health’s scenario translates to being observed toward the goal of health, in real time. The latter is particularly critical for heart conditions, where early notice can literally mean the difference between life and death. Sudden Cardiac Death (SCD) is still one of the top killers in human premature death. It is this latter fact that demands some real solution if we are going to be monitored outside of the hospital. Standard tracking systems are out of reach for the people who need them most, often in remote areas or underresourced regions such as a country or an unprofitable inner city. In this work, an experimental dataset of wearable ECG monitors in the smart glove with sensors is utilized. It's been able to monitor heart rates around the clock outside of a hospital, and in some cases where there is no doctor present at all. The main steps in the proposed procedure would be to gather data wirelessly into ECG sensors, perform real-time risk estimation at the edge and send (e.g., as Bluetooth message) its result for cloud processing backend. The uniqueness of our system is providing a community level emergency alerting system to the health care workers at risk with positions who are alerted in case of very high risk. Meanwhile, the reliability is guaranteed by evaluating performance measures such as detection accuracy, latency and energy consumption in this work. The results indicate that the proposed system yields high accuracy for detection and immediate response, and therefore constitute an efficient heart risk management tool in everyday life.

The workforce of today is significantly dependent on the internet. The frequency of hazardous online behavior is increasing along with the popularity of the internet. Identifying different security flaws in a computer network is the matter of concern. There are many different security breaches that can occur in a computer network, and intrusion detection system is the process of identifying them. An IDS makes sure that a network is safe. The CICIDS2017 dataset, which contains a novelty and variety of attacks, is used in this investigation to identify intrusions using decision tree and ensemble classification techniques. This research employs a genetic search based LMRI-HRCD feature selection on the dataset. Using proposed LMRI-HRCD feature selection method; the constructed system reduces the number of features as well as size of the dataset. The experimental results are analyzed using the machine learning application Weka. Research findings from empirical studies demonstrate that decision tree based ensemble methods results in higher accuracy in intrusion detection. The presented approach is efficient and performs better when decision tree and ensemble techniques are used with specific reduced features.

RNA interference (RNAi) has emerged as a promising strategy for controlling insect pests, offering precise and environmentally sustainable alternatives to traditional pest control methods. By introducing double-stranded RNA (dsRNA) that specifically targets essential genes involved in pest survival, RNAi disrupts gene expression in target organisms. Various delivery methods, including topical application, transgenic plants, and nanoparticles, have been developed to enhance the effective administration of dsRNA. However, RNAi faces several challenges, including off-target effects, species-specific variations in efficacy, and the potential for resistance development. Despite these obstacles, ongoing research is focused on addressing these issues and improving the efficiency of RNAi-based pest control approaches. Future prospects for RNAi in insect pest management include advancements in delivery technologies, the identification of novel target genes, and the integration of RNAi with complementary pest control strategies. In conclusion, RNAi represents a potential long-term and targeted solution for insect pest control. Continued research and technological advancements are driving its adoption and expanding its application in agriculture and forestry globally. This review provides an in-depth analysis of RNAi mechanisms, evaluates current strategies, highlights the challenges faced, and explores the future directions for its use in insect pest management.

Water pollution, particularly from floating debris such as plastics, has become a critical environmental issue, threatening aquatic ecosystems and biodiversity. Autonomous solutions for the detection and removal of waste are increasingly essential for maintaining water cleanliness and mitigating pollution. However, existing systems face limitations in real-time detection, accuracy, and adaptability to diverse aquatic environments. This paper utilizes the water pollution images dataset, comprising almost 300 high-resolution images from lakes, rivers, and coastal areas, representing various types of floating waste under different environmental conditions. In response to these challenges, this paper introduces an autonomous unmanned surface vehicle (USV) system equipped with the enhanced waste detection network (EWD-Net). EWD-Net improves upon traditional single-shot detection algorithms by integrating deeper feature extraction layers and enhancing computational efficiency, resulting in higher accuracy and faster detection. Additionally, the system includes the dynamic path optimization (DPO) module for efficient navigation and obstacle avoidance in complex water environments. The novelty of this system lies in its dual approach, combining advanced detection with optimized path planning, ensuring effective autonomous operation. The results indicate that the proposed model achieves an accuracy of 94.6%, outperforming existing algorithms and providing a robust solution for real-time waste detection and collection.

Sickle cell disease is a severe genetic blood disorder marked by the production of abnormal hemoglobin (HbS), leading to sickle-shaped red blood cells that obstruct blood flow and cause various problems, such as the increased risk of infection, persistent anemia, acute pain episodes, and organ damage. Roughly 100,000 Americans suffer from SCD, with approximately 40,000 of them being children. Black people have the highest frequency of the disease. There are six Food and Drug Administration (FDA)-approved drugs, hydroxyurea, L-glutamine, crizanlizumab- TMCA, voxelotor, Casgevy, and Lyfgenia, that are used for the prophylaxis and treatment of serious complications of sickle cell disease. Current treatment approaches focus on symptom management, including pain control, hydroxyurea to reduce pain crises, and transfusions for severe anemia. Based on the clinical trial results, L-glutamine and crizanlizumab-TMCA prevent cell damage and hemoglobin sickling by reducing the sickle cell crisis episodes. At the same time, voxelotor improves hemoglobin oxygen-binding capacity in patients with SCD. Novel therapies, such as gene therapy and clustered regularly interspaced short palindromic repeats/CRISPR-associated protein 9 (CRISPR-Cas9) technology, aim to correct the genetic defect. At the same time, stem cell and bone marrow transplants offer potential cures but are limited by the availability of donors and side effects. Ongoing research seeks to enhance treatment options and develop potential cures for SCD. This review attempts to present a comprehensive overview of the current therapeutic approaches and newly developed innovative medicines to combat and potentially eradicate SCD with an emphasis on their mechanisms, efficacy, and clinical implications.

Uterine Lipoleiomyoma: Case report and review of literature - SAJCRR- Print ISSN No: - Online ISSN No:- 2319-1090 Article DOI No:- 10.18231/j.sajcrr.2025.009, Southeast Asian Journal of Case Report and Review-South Asian J Case Rep Rev