Volume 13 | Issue 4

 Abstract

Maintaining discipline and security in educational institutions requires ensuring adherence to standard regulations. YOLOv11, OpenCV, and dlib are used in this study's AI-driven real-time Uniform Detection System to automatically identify and categorise people according to their uniform status. The device uses face recognition to identify pupils and confirm that they are wearing the necessary uniform. It also incorporates a video capture camera for ongoing surveillance. An automatic email notice is sent to the relevant authorities and pupils in the event that a uniform violation is found. A unique YOLOv11 dataset is used to train the algorithm, guaranteeing precise distinction between uniforms and casual attire. The device also flags any security issues by identifying unauthorised individuals. The system saves captured frames of infractions for later study and keeps detection logs with timestamps, uniform status, and facial recognition results to improve data management. Notifications are tracked via logging email alerts. OpenCV is used for real-time image processing, Roboflow is used for dataset preparation, and Google Colab is used for model training. This automated method guarantees adherence to institutional policies, improves security, and decreases the need for manual oversight. IoT-based remote alerts and dataset extension are examples of future advancements that will increase accuracy and dependability.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

AI surveillance, YOLOv11, uniform detection, facial recognition, OpenCV, real-time monitoring, student identification, automation, object detection, data storage, school security, compliance tracking, discipline enforcement.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The AI system called "Road Traffic Analysis Using YOLO-V4 & Deep Sort" uses advanced technology to automate vehicle tracking in real time. Increasing numbers of urban residents and cars place heavy burdens on current traffic management systems that operate slowly and have limits. This project solves video monitoring difficulties by using YOLO-V4 for object detection and Deep Sort for multi-object tracking to process both streaming and recorded videos. The system recognizes motor vehicles, gives those ID numbers, and follows them across several images to reveal road patterns and vehicle movement. The system uses improved processing tools such as non-max suppression and Kalman filter methods to detect objects reliably. The system uses these technologies together with Python and TensorFlow to work in all types of roads and smart city settings at top speed. This project lets transportation authority's control road activity better to ease congestion and better protect their users. Our project will keep improving through the addition of traffic flow prediction systems based on IoT technology plus cloud architecture for wider implementation. Most modern traffic systems would benefit from the addition of smart tolling functions plus automatic detection of crimes and self-driving car assistance as part of intelligent transportation research.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

YOLO-V4 , Deep Sort, Kalman Filter, Python , Tensorflow, Object Detection.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The Intelligent SMS Spam Classifier is a Flask web tool that uses a Multinomial Naive Bayes model to spot and block spam messages. Text preprocessing through CountVectorizer lets the system find spam accurately while identifying regular messages. More sophisticated rules are required to handle the growing number of spam messages which use phishing and fraud tactics. The project improves how well machine learning can detect spam by making it more flexible across different environments. The easy-to-use Flask framework lets people use their phones or PCs to classify messages live. The system can detect spam patterns from different kinds of text through its training on a wide variety of SMS data. Our system reliability increases through security features which include input protection checks and HTTP data encryption plus constant incoming message speed management. The application provides flexibility to use either local resources or cloud services such as Heroku and AWS as its deployment platforms. The system will gain more capabilities through support for different languages along with deep learning technology and API-based filtering systems with a mobile app addition.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

CountVectorizer, Intelligent SMS Spam Classifier, Naive Bayes Model.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The evolving e-commerce environment requires consumers to overcome product deal identification difficulties because prices consistently change among various platforms. Through web-based price comparison tool PriceMatch users can access real-time pricing data because of its use of advanced web scraping techniques data analytics alongside automation which extracts data from major online retailers Amazon, Flipkart and Croma. Users access knowledgeable purchasing decisions because the system tracks prices in real-time while providing historical data analysis and automatic price alert notifications. The application constructs its infrastructure from Flask and Selenium combined with BeautifulSoup thus delivering fast data retrieval and user-friendly operation. The system provides future-enhancing capabilities through its modular design structure while giving users options for making their selection between AI-assisted price predictions and mobile-based platform upgrades and broader marketplace reach expansion. The price comparison automation feature in PriceMatch provides shoppers with affordable shopping recommendations that eliminate repetitive work while improving their entire online shopping experience.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

E-Commerce Environment, PriceMatch, Amazon, Flipkart, Croma, BeautifulSoup.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The machine learning system called CardioMyx analyzes multiple medical parameters through its designed algorithm to detect heart disease risk in individuals. This system applies the Random Forest Classifier ensemble learning technique as a robust method to evaluate patient information which generates high-risk or low-risk categories. The model analyzes age, sex, cholesterol levels, blood pressure, and blood sugar together with lifestyle factors to predict precise risk ratings. Medical personnel gain essential diagnostic insights from advanced analytics built into CardioMyx which leads to fast medical interventions. The technical platform supports preventive cardiology decisions by letting doctors provide individualized medical treatments or lifestyle recommendations based on calculated risk outcomes. As extra educational data enters the model it becomes better at producing accurate risk results and maintains such precision across multiple patient populations. CardioMyx shows promise to change heart disease prognosis by cutting down human mistakes while supporting physicians to detect diseases early. Through its functions the system assists healthcare providers while allowing patients to understand their cardiac state which promotes active cardiovascular risk management.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

CardioMyx, Pressure, Blood Sugar, Cholesterol Levels, Heart Disease Prognosis.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The digital healthcare platform EPharma uses artificial intelligence together with machine learning capabilities to connect patients' symptoms detection to the generation of correct prescription recommendations. The Random Forest-based learning model in the system evaluates user symptoms to recommend medications effectively thus helping users avoid the dangerous implications of self-diagnosis and wrong self-treatment. EPharma uses strong authentication systems to guarantee safe user access between different user classes including patients and healthcare workers and pharmacists. The Flask framework creates the backend system and enables real-time data retrieval and maintains a MySQL database that operates with structured security parameters. The credibility of medical information in EPharma improves when users interact with external APIs such as Wikipedia and PubChem because this provides verified medication information along with their side effects and proper usage guidelines. Through its intuitive layout EPharma makes medical prescription tasks easy which lets healthcare reach more people with better operational efficiency. The system has built-in AI prediction technology which prevents medical mistakes and a security system using authentication protocols and encryption techniques protects confidential medical data.Through its medication management system EPharma assists healthcare providers with prescription automation which simultaneously decreases their workload while delivering better care outcomes to patients. Users of the platform can learn about their medication through educational resources which empower them with vital knowledge. The document explains how the system functions as well as describes installation methods and testing procedures while demonstrating its capabilities in modernizing healthcare operations. EPharma showcases how it changes digital healthcare by implementing intelligent automation with secure data handling in addition to improved user experiences while fixing current prescription system issues.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Epharma, Artificial Intelligence, Flask Framework, Wikipedia, Pubchem.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Agricultural disease management remains vital in the field because plant diseases directly affect crop production levels and they create threats for food safety together with economic stability. The conventional approach to plant disease diagnosis consists of agricultural experts conducting visual examinations which prove to be irregular yet lengthy and prevents adequate services to farmers who reside beyond normal accessibility. The Smart Agro-Cure project specifies the development of an AI-powered pesticide recommendation system based on Convolutional Neural Networks which function through image input. The system uses leaf images for automatic disease detection which then recommends appropriate pesticides to improve both accuracy and efficiency in accessible disease management operations. The deep learning model reaches high precision identification of diseases because it receives training on multiple plant disease varieties within its diverse dataset. A curated pesticide database within this system enables users to get green and optimal treatment suggestions. Image preprocessing methods including resizing, normalization and augmentation help the project reach higher accuracy because they enhance model generalization abilities. Through the Flask-based user-friendly graphical interface (GUI) system users can easily upload images to get immediate disease analysis. The software demonstrates excellent scalability together with deployment features that make it suitable for mobile applications and edge computing platforms to serve farmers worldwide.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Agricultural Disease, Crop Production, Agro-Cure, Convolutional Neural Networks.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Machine learning serves stock market prediction as an essential application through which people alongside organizations obtain data-driven investment intelligence. The system establishes stock market prediction through the implementation of machine learning approaches and deep learning techniques toward stock price forecasting. The system runs on Python while using Flask for web-based deployment which enables users to obtain real-time predictions through an effortless interface. The model acquires historical stock information from Yahoo Finance through the use of Pandas and NumPy and yfinance libraries for processing. Deep learning acceptance requires MinMaxScaler to normalize the data before deep learning processing. The system implements a Sequential model which applies Long Short-Term Memory layers and Dropout layers and Dense layers using TensorFlow Keras for effective prediction of temporal patterns. The system performs five stages as part of its operational scope which include the collection of data followed by preprocessing procedures and model development and evaluation before deployment. Additional possible enhancements to this design will integrate multi-stock forecasting together with sentiment analysis functionality and mobile device compatibility. The main mission is to establish a powerful deep learning algorithm that does stock forecast with superior precision yet the system will also showcase an easy-to-use interface and instant prediction capabilities for user-defined stocks.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Machine Learning, Investment Intelligence, Stock Market Prediction.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Thyroid nodules are abnormal growths in the thyroid gland that can be benign or malignant. Early detection and classification are crucial for effective treatment. This project utilizes Deep Learning, specifically Convolutional Neural Networks (CNN) with an extended VGG algorithm, to improve accuracy in thyroid nodule detection. The model processes ultrasound images to differentiate between benign and malignant nodules, reducing dependency on manual diagnosis.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Thyroid nodules, Deep Learning, Convolutional Neural Networks (CNN), VGG algorithm (extended VGG), Ultrasound images, Benign nodules, Malignant Nodules, Early detection, Accuracy improvement, Manual diagnosis, Medical imaging, and Artificial Intelligence (AI) in healthcare.

  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

???? ?????? ???? ??????? ?????? ??? ??? ?????? ??????? ??????? ???????? ???????

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Feminism is a social and political movement that advocates for the rights and equality of women. Thus, feminism from the start has been a mix of a movement and an ideology that seeks to acquire all kinds of equal rights for women in society. The history of feminism includes early feminist movements, such as the women's suffrage movement, which fought for the right of women to vote in political elections. The women's suffrage movement was a significant part of the broader feminist movement, as it sought to secure basic civil and political rights for women. First-wave feminism, a key part of the women's suffrage movement, focused on gaining political and legal rights for women, such as the right to vote, own property, and access education and employment opportunities. After the suffrage movement ended feminist movement took seat. Second wave feminism marks key aspects in the politics of Marxist and liberal feminist movement. After that, third wave feminism include intersectionality, inclusive, diversity, focusing on individuality and social justice. Fourth wave feminism have digital, inclusive and focusing on online activism, body autonomy and consent. Feminist theory provides framework analysing women oppression, challenging, patriarchal structure, and promoting gender equality. It encompasses various branches, including liberal, radical, Marxist, postcolonial, and intersectional feminism, to understand and address gender-based injustices.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Feminism, rights, equality and social justice.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Glucose monitoring device is designed to reveal the glucose tiers inside the blood constantly. This machine collects the facts from the person with the help of a sensor and this information is despatched to the Alerting and Advising device. a blood glucose take a look at is commonly accomplished by using piercing the skin (commonly, on the finger) to attract blood and this blood is then tested for the quantity of concentration of glucose through monitoring the glucose degree constantly. This increase in the values of glucose can be observed in boom in current. these current values are converted into voltage form by means of a consistent resistance and the calibrated table is cited in program of Alerting system. those voltage values from sensor are interfaced to the Nedelcu. even though the fee of the usage of blood glucose meters seems high, it's far believed to be a fee advantage relative to the prevented scientific costs of the complications of diabetes. The advantages consist of a discount inside the prevalence price and severity of long-term complications from hyperglycaemia in addition to a reduction inside the brief-time period, probably existence-threatening complications of hypoglycaemia.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords - Glucose Monitoring, Electrochemical Sensors, Diabetes Management, Continuous Glucose Monitoring (CGM), IoT-based Monitoring, Real-time Data Processing, Machine Learning Predictions, Wireless Data Transmission.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper outlines the design, development, and installation of a solar-powered advertisement board as an environmentally friendly and cost-effective means of replacing electric-powered signage. As the demand for energy-efficient solutions increases and the price of electricity continues to rise, it has become vital to incorporate renewable energy sources in commercial uses. This project revolves around using solar energy to drive advertisement boards and minimize the use of grid electricity and reducing environmental footprint. The system comprises photovoltaic (PV) panels for generating energy, a charge controller for effective power management, a battery storage device for sustained operation, and an LED screen for content display. Energy efficiency, ruggedness, and cost-effectiveness are the key considerations in the design to guarantee sustained usage. By conducting extensive tests and analysis, the performance of the system is quantified under varying environmental conditions. The findings reveal that solar-powered advertisement boards can function efficiently with little maintenance, and they are a viable option for contemporary advertising purposes. This research is an addition to the development of green technology in digital advertising, encouraging the use of renewable energy for commercial and public display applications

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

solar energy, advertisement, LED display

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Abstract: The Advanced Encryption Standard (AES) is a widely used symmetric key cryptographic algorithm ensuring secure data transmission. This paper presents an optimized AES implementation using Verilog HDL, enhancing both performance and resource efficiency. The design incorporates fundamental AES operations, including SubBytes, ShiftRows, MixColumns, and AddRoundKey, while employing optimization techniques to reduce latency and hardware utilization. A key enhancement is the reuse of the S-box for both encryption and decryption, with a multiplexer selecting between the two modes. Additionally, the MixColumns and AddRoundKey operations are integrated into a single unit to streamline processing. The proposed design further improves efficiency by reducing S-box instances through optimized key expansion, generating a 128-bit key every five cycles. Security is reinforced by extending the key size to 256 bits, with unique transformations applied across different encryption rounds. Functional simulation and synthesis results validate the correctness and effectiveness of the implementation, demonstrating its suitability for resource-constrained environments such as IoT devices and FPGA-based secure communication systems. This study highlights the advantages of Verilog HDL in cryptographic hardware design, offering scalability and flexibility for future advancements in encryption techniques.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

AES, Cryptography, Verilog HDL, S-box optimization, FGPA

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The increasing influence of social media has turned into a defining feature of present day college experience, particularly in Bangladesh, where platforms like Facebook, Instagram and WhatsApp are deeply integrated into students' daily routines. While social media offers various benefits, such as facilitating connections, access to knowledge, and emotional support, its excessive consumption has been tied to the surge in anxiety level among college students. This article explores the dual impact of social media on anxiety among Bangladeshi college students, examining both the negative and positive aspects. On one hand, social media fosters social comparison, cyberbullying, and digital addiction, all contributing to heightened stress and anxiety. On the other hand, it also provides opportunities for creative expression, mental health support, and community-building. The article further discusses the cultural and academic pressures that intensify the anxiety students experience, emphasizing the need for mindful social media use. The article concludes with recommendations for students to adopt healthier social media habits and for institutions to recognize the importance of mental health awareness.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Social Media, Anxiety, Facebook, Instagram, and WhatsApp, YouTube and TikTok

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Concerns have been raised about the possibility that the criminalization of medical negligence could serve as both a threat to the medical profession and a deterrent in the healthcare industry. When a healthcare professional fails to provide patients with the expected level of care, this is known as medical negligence. Historically, civil litigation was the primary means of addressing medical negligence; however, the growing trend toward criminalizing such conduct aims to provide a more stringent accountability mechanism. Critics contend that criminalizing healthcare providers is a necessary deterrent because it ensures that they are held accountable for their actions and upholds high standards of care, increasing patient safety and trust in the medical system. Critics, on the other hand, warn that making medical negligence a criminal offense could create a climate of fear and hesitation among medical professionals, which could lead to defensive medicine and lower quality care. Open communication, professional autonomy, and existing problems like physician burnout could all be exacerbated by the threat of criminal prosecution. The ramifications of criminalizing medical negligence are examined in this abstract, with the purpose of determining whether or not it poses a threat to the integrity and well-being of the medical profession or merely serves as an effective deterrent. In order to determine whether or not this strategy is effective and fair, it is essential to strike a balance between accountability and the maintenance of professional trust.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Negligence, Medical, Harm, Proffessional, Liability, Care, Patient, Criminalisation

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Cross-Lingual Sentiment Analysis (CLSA) has emerged as a vital instrument for comprehending public sentiment around the world as digital communication crosses language and geographic barriers. A thorough analysis of CLSA approaches, such as translation-based models, multilingual NLP strategies, and cross-lingual transfer learning, is presented in this research. We go over important issues including linguistic diversity, the scarcity of annotated datasets, and cultural differences in attitude. Lastly, we examine new prospects such as the creation of multilingual sentiment-aware AI systems, improved machine translation for sentiment preservation, and improvements in huge language models. The importance of cross-lingual sentiment analysis (CLSA) in a multilingual digital world is examined in this research. We go over the CLSA methods, the main language translation issues, cultural variations, and data accessibility. We also look at new approaches and potential paths forward to enhance sentiment analysis in various languages. The increasing requirement to analyses feelings in a multilingual, diversified global context has made cross-lingual sentiment analysis a crucial component of contemporary natural language processing (NLP). This work offers a thorough examination of cross-lingual sentiment analysis, examining the difficulties and methods employed to overcome linguistic and cultural differences in sentiment classification. To improve sentiment analysis across languages, the study focuses on the current approaches, the difficulties of working with low-resource languages, and the use of machine translation, transfer learning, and multilingual models. Furthermore, by using creative methods for data gathering, model adaption, and cross-lingual transfer, we highlight new trends and chances to raise the calibre of cross-lingual sentiment analysis.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Cross lingual sentiment analysis, NLP

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Jhumpa Lahiri's The Lowland explores the complexities of cultural identity, displacement, and acculturation through the experiences of its characters, particularly Bela, the daughter of Indian immigrants. The novel delves into themes of cultural estrangement and hybridity, examining how Bela navigates her dual heritage as a second-generation immigrant in the United States. Drawing on W.E.B. Du Bois's concept of double consciousness, the analysis highlights Bela's internal conflict as she struggles to reconcile her Indian roots with her American upbringing. Her fragmented identity is further compounded by her mother Gauri's abandonment and her father Subhash's emotional distance, leaving her culturally adrift. The novel also embodies postmodern themes, such as fragmented narratives and the rejection of traditional structures, as seen in Bela's rootless existence and her defiance of familial and societal expectations. Through Bela's journey, Lahiri underscores the fluidity of identity in a globalized world, where cultural belonging is neither fixed nor singular. The study concludes that The Lowland portrays identity as a mosaic of experiences, shaped by dislocation, multicultural influences, and the absence of cohesive cultural anchors.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Cultural Estrangement, Acculturation, Hybridity, Postmodernism, Displacement, Double Consciousness

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Accurate demand forecasting is critical in the food industry to minimize waste and optimize inventory management, particularly for perishable products. This study uses the "Food Demand Forecasting" dataset by Genpact to compare 16 forecasting models, including traditional machine learning (Random Forest, Gradient Boosting, LightGBM, XGBoost, CatBoost, ElasticNet, SVR), statistical methods (Prophet, TBATS, ARIMA), deep learning models (LSTM, BiLSTM, TCN, ETSformer, GNN), and a hybrid LSTM + XGBoost approach. We applied advanced feature engineering, such as lag features, Exponentially Weighted Moving Average (EWMA), and cyclical encoding of temporal features, to capture seasonality and trends. The primary focus was on achieving high accuracy, with the Hybrid LSTM + XGBoost model achieving the highest accuracy of 93.47%, followed closely by the BiLSTM model at 91.84%. The TCN model achieved a competitive accuracy of 90.66%, demonstrating its effectiveness in capturing temporal patterns. This study improves upon prior work by Panda and Mohanty (2023) by introducing a broader model set, explicit seasonality modeling, and a focus on maximizing prediction accuracy. We provide detailed reasoning for model selection, feature engineering, and evaluation metrics, emphasizing the practical implications of high accuracy in food supply chain management.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Machine Learning, Artificial Intelligence , Time Series Analysis

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Women-owned businesses are in minority in India. The hurdles faced by women who have embraced entrepreneurship are vast and often different from those experienced by their male counterparts. Karnataka has always been a supporting ecosystem that has facilitated the growth of innovation and entrepreneurship. Universally, entrepreneurship is accepted as the main driver of profitable growth. It has appeared as an employment outlook and a way of adding women at diverse levels within an economy. Women entrepreneurs are gradually becoming one of the elements of growth, predominantly in India. Karnataka is one of the top five states in the whole of India, which is housing the maximum number of women entrepreneurs in the nation with a majority in small and medium-sized businesses. The influence that entrepreneurship has on the well-being of women is not sufficiently discovered.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

entrepreneurship, Women entrepreneurs, Kalaburagi city, motivation, E-commerce

  License

Creative Commons Attribution 4.0 and The Open Definition