Volume 14 | Issue 4 |

 Abstract

Agriculture remains a primary livelihood for a large portion of the global population; however, farmers continue to face challenges such as price uncertainty, market fluctuations, and limited trust in existing trading systems. In many cases, traditional supply chains depend on intermediaries and centralized databases, which can reduce farmers' profit margins and restrict transparency. Issues such as data inconsistency and lack of reliable market information further complicate decision making. This paper presents Krishi-Unnati, a digital platform designed to improve transparency and support better decision making in agricultural trading. The system integrates blockchain technology, artificial intelligence, and mobile computing within a hybrid architecture. Transactional data, including trade records, is stored on a blockchain ledger to ensure integrity, while other operational data is maintained off-chain to improve efficiency. The platform provides two main functionalities. First, a crop price prediction module based on Random Forest Regression is used to estimate future market prices. Second, a buyer recommendation engine analyzes historical transaction data to suggest suitable trading partners. Together, these components assist farmers in making more informed choices regarding crop sales and market participation. The mobile application is developed using React Native, with a backend built on Node.js, Express.js, and MongoDB. System evaluation indicates improvements in transparency, access to market information, and overall decision support. The proposed approach contributes toward a more reliable and data-driven agricultural trading ecosystem.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Blockchain, Smart Agriculture, Price Prediction, Buyer Recommendation, Supply Chain Transparency, Smart Contracts, Agriculture Technology

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The increasing integration of Artificial Intelligence (AI) into cyber security systems has transformed the way digital threats are detected, analyzed, and mitigated. While AI enhances efficiency and accuracy in cyber defense mechanisms, it simultaneously introduces complex legal and ethical challenges. This paper critically examines the implications of AI-driven cyber security, focusing on issues such as data privacy, algorithmic bias, accountability, and lack of transparency. AI systems rely heavily on large datasets, raising concerns regarding unauthorized data collection and potential misuse. Moreover, the opaque nature of AI algorithms often complicates legal accountability when automated decisions result in harm. Ethical concerns such as excessive surveillance, discrimination, and erosion of individual rights further intensify the debate. Existing legal frameworks remain inadequate to fully regulate AI technologies, highlighting the urgent need for comprehensive and adaptive policies. This study emphasizes the importance of balancing technological advancement with ethical responsibility and legal compliance. It concludes that a multidisciplinary approach involving legal experts, technologists, and policymakers is essential to ensure that AI-driven cyber security operates within a framework that respects human rights and promotes transparency, fairness, and accountability.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial Intelligence, Cyber Security, Legal Challenges, Ethical Issues, Data Protection, Algorithmic Bias

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

ekuokf/kdkj] Ekkuokf/kdkj guu] lkekftd vlekurk] dkuwu izorZu] cky Je] efgyk fgalk] xjhch] HkkjrA

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Rapid urbanization in India has overwhelmed traditional, reactive complaint systems, leaving ward-level issues like garbage overflow and infrastructure decay unresolved. This paper proposes WardPulse, an AI-driven proactive monitoring system integrating IoT sensors, Computer Vision, and crowdsourced reporting. By utilizing Convolutional Neural Networks (CNN) with MobileNetV2 for image-based defect detection and a geo-tagged dashboard for municipal authorities, the system automates issue classification and prioritization. Experimental results show a 40% reduction in resolution time and significant improvements in citizen engagement. This scalable framework is designed for seamless deployment in Smart City initiatives across both municipal corporations and gram panchayats.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Smart Cities, Computer Vision, MobileNetV2, IoT, Civic Technology, and Urban Infrastructure.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Tomato (Lycopersicum) is one of the most economically important horticultural crops in India, especially in Andhra Pradesh, which accounts for about 18% of the country's tomato production. Tomato plants are greatly affected by many different leaf diseases. These include fungal problems like Early Blight and Septoria Leaf Spot, bacterial issues such as Bacterial Spot, and viral infections like Yellow Leaf Curl Virus and Mosaic Virus. If these diseases are not found and dealt with quickly, they can cause a loss of 40 to 60% of the crop. Traditional ways of identifying diseases by looking at plants, done by agronomists, take a lot of time, are easy to make mistakes, and don't work well when you need to check a large number of crops, especially in rural areas where it's hard to get expert help. This project introduces an automated tomato leaf disease identification system utilizing the Inception V3 deep convolutional neural network architecture combined with transfer learning. The model was trained using the PlantVillage dataset, which includes 10 different categories of tomato leaves and around 18,160 images, all obtained from Kaggle. A two-step transfer learning approach is used: first, features are extracted using the frozen weights from the pre-trained ImageNet model, and then the top 30 base layers are fine-tuned with a lower learning rate. Data augmentation techniques such as horizontal and vertical flipping, random rotation, zoom, and brightness adjustment are utilized to tackle class imbalance and enhance model generalization. The system gets an average accuracy of 97.2% on the test set it hasn't seen before. It also has a precision of 0.923, recall of 0.916, and an F1-score of 0.919 when looking at all classes equally. The trained Keras model is converted into ONNX format to make CPU inference run 30 to 40% faster using ONNX Runtime. An HSV-based leaf detection module filters out non-leaf inputs before making predictions, and it uses a confidence threshold to clearly mark any uncertain results

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Tomato Leaf Disease Detection, Inception V3, Transfer Learning, Convolutional Neural Network (CNN), PlantVillage Dataset, Deep Learning, Data Augmentation, Precision Agriculture

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Cardiovascular disease (CVD) is one of the major causes of mortality around the globe, necessitating the need for early detection that will enable better health care results. Existing diagnostic methods utilize significant clinical expertise along with extensive testing; hence, it becomes time-consuming in detecting diseases. This study proposes the "Deep-Ensemble Cardiovascular Disease Detection System" which uses the power of deep learning to detect CVDs at an early stage. The proposed system incorporates a number of deep learning algorithms to develop an ensemble for better performance in predicting the results. In the data preprocessing process, the dataset is cleaned and balanced to improve model learning. To ensure effective prediction and detection of heart diseases, the ensemble uses the CNN and GRU architectures to capture the patterns from health data. Primarily, rather than classifying patients in two categories, including No Disease and Disease, the categorical prediction is extended to four categories, including No Disease, Low Risk, Medium Risk and High Risk. Also, this categorical prediction is not limited to two categories, namely No disease and Disease but extended to four categories, namely No Disease, Low Risk, Medium Risk and High Risk. First, instead of two categories of predictions, such as No Disease and Disease, the categorical prediction is expanded to four categories, including No Disease, Low Risk, Medium Risk, and High Risk. Second, it does not only involve the prediction of two categories, i.e., No disease and Disease, but the expansion to four categories, including No Disease, Low Risk, Medium Risk, and High Risk. Third, an online cardiovascular disease prediction and an internet-based application are realized. Lastly, a categorization of risk prediction is performed by the system. From experimental results, it can be observed that the proposed deep-ensemble model assures higher levels of accuracy, reliability, and usability compared to conventional machine learning methodologies. The proposed system can be used by healthcare professionals for the purpose of early diagnosis, risk assessment, and preventive treatment planning, which helps improve patient care and reduce mortality.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Cardiovascular Disease Detection, Deep Ensemble Learning, CNN-GRU Model, Multi-Class Classification, Risk Prediction, Web-Based Prediction System, Artificial Intelligence, Healthcare Analytics.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The basic thesis of growth is the economic growth of which is required for political, social and economic stability the quality environment normally assumes lower priority in planning proposals and long-term planning. Unlimited exploitation of nature by man disturbed the ecological balance between living and non-living components of the biosphere. The adverse conditions created by man himself threatened the survival not only of man himself but also other living organisms. Due to progress, industries, technology, chemicals, atomic energy, there are a number of industrial effluents and emissions of poisonous gases in the atmosphere and also added solid waste which has lowered the quality of environment. The pollution is a necessary evil for all development. Due to lack of development of culture of pollution control, there has resulted a heavy backlog of gaseous, liquid and solid pollution in our country. Thus, pollution control in our country is a recent environmental concern. There is a race in developed countries to exploit every bit of natural resources to convert them into goods for their use and comfort and to export them to other needy countries. The industrialized countries dump lot of materials in their environment which becomes polluted. The environmental pollution has lowered its quality.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Environmental Degradation, Pollution, Institutional Failure, Conflicts, Energy

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The increasing prevalence of skin diseases worldwide highlights the need for early and accurate diagnosis to prevent complications and ensure timely treatment. Traditional diagnostic approaches rely heavily on dermatologists' expertise and manual examination, which can be time-consuming, subjective, and limited in accessibility. To address these challenges, this project proposes an AI-assisted skin disease diagnosis system integrated with a virtual medical expert and appointment booking facility, leveraging deep learning and advanced image processing techniques to provide efficient, accurate, and user-friendly healthcare support. The system enables users to upload skin images, which are preprocessed using techniques such as denoising, resizing, segmentation, and normalization to enhance image quality. Deep learning models, including Convolutional Neural Networks (CNN), ResNet, and EfficientNet, are utilized to classify a wide range of skin diseases with high accuracy. The model is trained and validated on diverse datasets to ensure robustness and reliability in real-world scenarios. Upon diagnosis, the system provides detailed results along with precautionary measures and basic medical recommendations. To enhance user interaction and accessibility, a virtual medical expert chatbot is integrated into the system. This chatbot offers explanations of predicted conditions, suggests preventive steps, and answers user queries, thereby bridging the gap between patients and medical professionals. Furthermore, the system incorporates an appointment booking feature that allows users to schedule consultations with dermatologists, ensuring a seamless transition from diagnosis to treatment. In conclusion, the proposed system presents a comprehensive solution by combining AI-based diagnosis, virtual consultation, and appointment scheduling into a single platform. Future enhancements may include expanding datasets for better generalization, improving model interpretability through explainable AI, and ensuring data privacy and compliance with healthcare standards. This project demonstrates the potential of integrating artificial intelligence with healthcare services to deliver more accessible, efficient, and patient-centered solution.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

CNN, ResNet, EfficientNet

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Financial awareness is essential in influencing the investment habits of retail investors. Based on theories like the Financial Literacy Theory and the Theory of Planned Behavior, this research describes how awareness, mindset, and perceived control affect investment choices. Through government-supported initiatives like the RBI Retail Direct Scheme, individuals can engage directly in government securities markets without needing intermediaries. This research explores how financial awareness influences retail investors in Bengaluru to adopt government direct retail investment schemes. The study relies on primary data gathered from 120 participants through a structured questionnaire. Statistical methods like percentage analysis, correlation, and regression were used to examine the data. The results show a strong positive correlation between financial awareness and the uptake of these schemes, reinforcing the theoretical notion that increased knowledge results in improved financial decision-making. Nonetheless, elements like insufficient awareness, intricate procedures, and restricted digital literacy persist as obstacles. The research highlights the importance of focused financial education programs, streamlined investment procedures, and digital literacy initiatives to enhance participation in government-supported investment platforms.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Financial Awareness; Financial Literacy; Retail Investors; Investment Behaviour; Government Securities; RBI Retail Direct Scheme; Financial Inclusion; Digital Financial Literacy; Investment Decision-Making; Investor Awareness

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

ABSTRACT:- India is a Socialist Democratic country and its economy is based on agriculture. Land is one of the most important sources of income. Human beings has right to live this land but the ownership of the land varies. Since the early stage did not have any law for the ownership of the land but the time passed then scenario was changed. The legal perspective of land acquisition in India has undergone a remarkable transformation with the exchange of the colonial Land-Acquisition Act of 1894 by the Right to Fair Compensation and Transparency in LARR, 2013.This paper tried to establish the development of land acquisition laws in India and the transformation of new LARR Act. Through an analysis of statutory provisions, judicial interpretation, the study highlights both progressive feature of the act and the persistent gaps in its execution.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

KEY WORDS: -Land, Acquisition, Transformation, Compensation, Legal.

  License

Creative Commons Attribution 4.0 and The Open Definition