Volume 13 | Issue 4 |

 Abstract

Face recognition is a powerful and widely adopted biometric technology that allows systems to automatically identify or verify an individual based on facial features. In an era where security is a major concern, face recognition presents a noncontact and highly efficient method for personal identification. Project explores a face recognition system developed using Python and OpenCV. The system detects, stores, trains, and identifies human faces by capturing and analyzing facial data. A webcam is used to collect facial images of users, which are later used to recognize the individual in real-time.. The is simple which helps non technical user to use .Project demonstrates how artificial intelligence and image processing can work together to improve digital security and ease the authentication process in real-life scenarios like attendance tracking, access control, and digital verification.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Face Recognition , Face Detection , attendance system.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Prosthetic technology has undergone a significant transformation with the advent of intelligent systems that integrate bio-signal processing, machine learning, and real-time control mechanisms. This review paper presents an in-depth exploration of a Smart Bionic Hand system that combines low-cost hardware components such as servo motors, Raspberry Pi, Arduino microcontrollers, and various sensors (EMG, flex, and gyroscopic sensors) with artificial intelligence algorithms to enable intuitive, adaptive, and affordable prosthetic solutions. The system captures electromyographic (EMG) signals from the user's muscles, interprets them using AI models, and actuates the mechanical hand to mimic natural human gestures. A feedback loop ensures real-time response and system learning, offering a high level of customization and comfort for the user. This approach addresses the shortcomings of traditional prosthetics, including high cost, lack of feedback, and poor adaptability. The review discusses system architecture, literature background, implementation details, and analytical performance of the Smart Bionic Hand in real-world scenarios, alongside future improvements.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords: Smart Bionic Hand, Prosthetics, Electromyography (EMG), Artificial Intelligence, Adaptive Control, Raspberry Pi, Gesture Recognition, Low-cost Design, Bio-mechatronics, Real-time Feedback

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Distributed network attacks are referred to, usually, as Distributed Denial of Service (DDoS) attacks. These attacks take advantage of specific limitations that apply to any arrangement asset, such as the framework of the authorized organization's site. In the existing research study, the author worked on an old KDD dataset. It is necessary to work with the latest dataset to identify the current state of DDoS attacks. This paper, used a machine learning approach for DDoS attack types classification and prediction. For this purpose, used Random Forest and XGBoost classification algorithms. To access the research proposed a complete framework for DDoS attacks prediction. For the proposed work, the UNWS-np-15 dataset was extracted from the GitHub repository and Python was used as a simulator. After applying the machine learning models, we generated a confusion matrix for identification of the model performance. In the first classification, the results showed that both Precision (PR) and Recall (RE) are _89% for the Random Forest algorithm. The average Accuracy (AC) of our proposed model is _89% which is superb and enough good. In the second classification, the results showed that both Precision (PR) and Recall (RE) are approximately 96% for the XGBoost algorithm. The average Accuracy (AC) of our suggested model is 96%. By comparing our work to the existing research works, the accuracy of the defect determination was significantly improved which is approximately 85% and 79%, respectively.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

CNN(Convolutional Neural Network), LCNN(Lookup based Convolutional Neural Network), RNN(Recurrent Neural Network), DEX(Dalvik Executables), TCP(Transmission Control Protocol), IP(Internet Protocol), HTTP(Hyper Text Transfer Protocol), ADT(Android Development Tool).

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Predicting water quality is essential for ensuring public health and sustainable water resource management. This study explores the application of machine learning algorithms, specifically Random Forest (RF) and Naive Bayes (NB), for effective water quality prediction. Using a dataset composed of various physicochemical parameters, we analyze and classify water quality indicators to assess its suitability for consumption and environmental health. Random Forest, an ensemble learning method, is leveraged for its robustness in handling large datasets and its ability to capture complex patterns in water quality features. Naive Bayes, a probabilistic classifier, complements this by providing a simple yet effective approach to classify water quality based on conditional probabilities. Both models are evaluated in terms of accuracy, precision, recall, and F1-score, with comparative analysis to highlight their strengths and limitations. The results demonstrate that combining the predictive accuracy of Random Forest with the interpretability of Naive Bayes offers a practical approach for water quality monitoring, supporting real-time decision-making and regulatory compliance in water resource management.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

These days people can't live without using social media either in personal life of individual or in the academics. All we accept that it is the contemporary technological era, where social media has emerged as a powerful influence across various spheres, including higher education. This study investigates the usage patterns of social media among teachers and students in Higher Education Institutions (HEIs) in West Bengal, India. With the increasing integration of platforms such as Facebook, YouTube, and live-streaming tools into academic practices, it is vital to understand both the opportunities and challenges they present. Utilizing a descriptive survey method, the study sampled 40 teachers and 200 students across five universities using multistage random sampling. Data were collected through two distinct questionnaires developed for teachers and students. Findings reveal that social media is predominantly used for educational purposes, communication, recreation, and encouraging social responsibility. Notably, the COVID-19 pandemic accelerated the shift in perceptions, positioning social media as a critical tool for sustaining education during crises. However, concerns such as privacy risks, distraction, and ethical issues also surfaced. The study underscores the need for strategic policies and training programs to maximize the educational benefits of social media while mitigating its drawbacks. This study may contribute to the growing body of knowledge on digital integration in higher education specially teacher education and offers valuable insights for educators, policymakers, and students aiming to navigate the digital learning environment more effectively.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Social Media, Teachers, Higher Education Institutions, Facebook, Instagram, Telegram, X, YouTube, WhatsApp

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Ayodhya, a city of profound religious and cultural significance, has emerged as a prominent destination for religious tourism in India. Known as the birthplace of Lord Rama and home to the recently inaugurated Ram Janmabhoomi Temple, Ayodhya has witnessed a rapid transformation driven by spiritual, historical, and cultural narratives. This study explores the development of religious tourism in Ayodhya from a socio-economic perspective, highlighting its impact on local communities, infrastructure, employment generation, and economic diversification. The research examines how religious tourism contributes to income opportunities, revitalizes traditional livelihoods such as handicrafts and hospitality, and fosters cultural preservation. Simultaneously, it addresses the challenges posed by rapid urbanization, environmental pressures, and socio-cultural shifts. By analysing the interplay between religious heritage and socio-economic development, the study underscores the need for sustainable tourism planning those balances economic growth with cultural integrity and community welfare. The findings aim to inform policy frameworks and development strategies to ensure inclusive and long-term benefits from Ayodhya's growing religious tourism sector.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Religious Tourism, Socio-economic perspective, Cultural integrity, Historical Background and Ayodhya Municipal Corporation.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

System presents a hybrid movie recommendation system designed to merge collaborative filtering, content-based filtering, and cosine similarity, offering users personalized suggestions rooted in their preferences and viewing history. Built as a web application with an HTML/CSS frontend, the system dynamically retrieves movie data via APIs to circumvent static dataset limitations. User engagement is heightened through visual comparisons of watched and recommended content. Sentiment analysis of reviews, implemented using Support Vector Machines (SVM), further refines recommendation accuracy. By integrating collaborative and content-based methods, the system addresses challenges like data sparsity and the cold start problem. Future plans include transitioning the platform to Flutter for improved interactivity and mobile compatibility. System underscores the efficacy of hybrid models in enhancing recommendation diversity and user satisfaction.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

hybrid recommendation system, collaborative filtering, content-based filtering, sentiment analysis, Api integration, mobile adaptation

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The rapid expansion of the Internet of Things (IoT) has enabled transformative applications across various sectors such as healthcare, smart cities, and industrial automation. However, this surge in connectivity has simultaneously exposed IoT networks to heightened cybersecurity threats, particularly Distributed Denial of Service (DDoS) attacks. Due to limited processing and security capabilities, IoT devices are easily compromised, making traditional Intrusion Detection Systems (IDS) inadequate in such environments. This study introduces an adaptive and lightweight IDS framework that utilizes a hybrid ensemble of bio-inspired optimization techniques--Spotted Hyena Optimizer (SHO), Parrot Optimizer (PO), and Grey Wolf Optimizer (GWO)--for feature selection. By embedding a self-attention mechanism, the system dynamically identifies key features that enhance detection accuracy while minimizing computational costs. The selected features are used to train machine learning classifiers including Decision Tree, SVM, Random Forest, XGBoost, and shallow Neural Networks. Evaluated on the UNSW-NB15 dataset, the proposed model demonstrates high performance with reduced false positives and latency, offering a scalable and real-time solution for DDoS mitigation in IoT ecosystems.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Internet of Things (IoT), Distributed Denial of Service (DDoS), Intrusion Detection System (IDS), Bio-Inspired Optimization, Feature Selection, Machine Learning.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Abstract: With the rapid advancement of wireless communication technologies, 5G mm Wave patch antennas have become increasingly important due to their capability to operate at high frequencies, their small size, and their support for high-speed data transmission. However, the development and analysis of these antennas often involve time-consuming full-wave electromagnetic simulations, which can slow down the design cycle. To overcome this limitation, Artificial Neural Networks (ANNs) offer a promising alternative by enabling quick prediction of essential antenna characteristics. In this work, a compact patch antenna operating in two bands 38.93-39.63GHz and 41.79 GHz to 42.92 GHz band is introduced, making it well-suited for 5G applications. The trained ANN model provides fast and accurate S11 predictions, effectively minimizing the need for lengthy simulation processes.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial neural network, 5G antenna, Machine Learning

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

dakshin-purv asia mein bharat ki pahuch:act east niti ki uplabdhiyan aur chunautiyan

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

dakshin-purv asia mein bharat ki pahuch:act east niti ki uplabdhiyan aur chunautiyan

  License

Creative Commons Attribution 4.0 and The Open Definition