Volume 14 | Issue 4 |

 Abstract

Thuja occidentalis L. is a species of the Cupressaceae family that has been widely researched as regards its aerial part, most commonly leaves and twigs. Nevertheless, one of the least explored roots is the root system even though it has been used in traditional medicine in many indigenous and folk practices. This review summarizes the existing information on the phytochemical composition of the root, pharmacognostic, ethnobotanical, extract, and pharmacological properties. Particular emphasis is put on the comparison of root and aerial parts chemistry and bioactivities. The root has a different chemical profile with reduced thujone and increased phenolics and sesquiterpenes. Adventitious root morphology and resin duct anatomy are pharmacognostic features that are described. Aerial part uses are opposed to conventional uses of digestive and genitourinary disorders. Improved extraction methods to include Soxhlet, ultrasound-assisted and green solvent methods have been shown to yield better bioactivities. Antimicrobial, anti-oxidant, and anti-inflammatory potential are examples of biological activities that help to support ethnomedicinal claims. In spite of the insights, there is still a gap in the research between standardization, mechanistic research and clinical validations, which underscores the root as an untapped therapeutic resource that is subject to further research.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Thuja occidentalis, phytochemistry, pharmacognosy, traditional medicine, extraction methods, bioactive compounds, comparative analysis.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This study explores the influence of brand labels on consumer trust in India's food market. In an environment characterized by increasing consumption of packaged foods and rising concerns regarding safety and authenticity, brand labels act as critical signals that guide consumer decision-making. The research integrates established branding theories, including Aaker's Brand Equity Model and Keller's Customer-Based Brand Equity framework, to examine how elements such as brand names, visual design, and labeling cues contribute to trust formation. Primary data was collected through a structured questionnaire administered to respondents across major urban centers, including Delhi, Mumbai, and Bengaluru. The study evaluates key variables such as brand perception, label interpretation, trust indicators, and purchase behaviour. Statistical tools including descriptive analysis and regression are employed to assess relationships between label attributes and consumer trust. The findings are expected to demonstrate that effective brand labeling significantly enhances consumer trust, which in turn influences purchase intention and brand loyalty. The research contributes to the existing literature by providing insights specific to the Indian food sector and offers practical implications for marketers aiming to build credible and trustworthy brand identities.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Brand Labels, Consumer Trust, Food Market, Packaging, Brand Equity, Purchase Behaviour

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Automated road divider detection is a crucial component in intelligent transportation systems and autonomous driving applications. This project presents a real-time detection system using the advanced deep learning model YOLOv12 (You Only Look Once). The system is designed to accurately identify road dividers from images and video streams under varying environmental conditions such as lighting changes, shadows, and traffic density. The proposed approach leverages computer vision and object detection techniques to process input data, detect road dividers, and highlight them with bounding boxes. YOLOv12 enables high-speed and efficient detection by analyzing the entire image in a single pass, making it suitable for real-time applications. The model architecture is optimized to balance accuracy and computational efficiency, ensuring effective performance even on resource-constrained devices. The dataset used for training includes diverse road scenarios such as highways, urban roads, and rural environments, which improves the robustness of the system. Data augmentation techniques such as rotation, scaling, and brightness adjustment are applied to enhance model generalization. The system is evaluated using performance metrics like precision, recall, and mean Average Precision (mAP) to ensure reliability. Experimental results indicate that the proposed system achieves high detection accuracy with low latency, making it suitable for practical deployment. Additionally, the model demonstrates strong performance in detecting partially visible or occluded road dividers. This solution can be integrated into advanced driver assistance systems (ADAS), traffic monitoring systems, and autonomous vehicles to enhance road safety and navigation efficiency. Furthermore, the system can be extended by incorporating additional features such as lane detection, obstacle detection, and real-time alert mechanisms for drivers. With continuous improvements and integration of emerging technologies, automated road divider detection has the potential to play a significant role in the development of safer and smarter transportation systems.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Automated Road Divider Detection, YOLOv12, Object Detection, Computer Vision, Deep Learning, Intelligent Transportation Systems, Autonomous Vehicles, Real-Time Detection, Image Processing, Bounding Box Detection, Traffic Monitoring, ADAS (Advanced Driver Assistance Systems), Convolutional Neural Networks (CNN), Road Safety, Smart Transportation, Video Analysis, Feature Extraction, Model Training, Dataset Annotation, Edge Computing, Embedded Systems, Scene Understanding, Urban Traffic Management,

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This paper proposes an intelligent Career Bridge Platform that bridges the gap between academic education and industrial needs by offering personalized skill assessment, learning paths, and career recommendations. The architecture of the platform is composed of a FastAPI backend with PostgreSQL database management for handling users' profile information, job history, and technical knowledge. The proposed framework utilizes Google Gemini (Generative AI) to assess users' skills and knowledge based on resume data, GitHub portfolio, LinkedIn profile, and eLearning records. From the assessment results, the Career Bridge Platform offers adaptive technical quizzes and constructs a personalized learning roadmap consisting of Foundation, Intermediate, and Advanced stages. The framework also incorporates the use of Adzuna Job Search API to offer users real-time job recommendations based on their domain, competencies, and learning roadmap topics. A mixed job matching method that employs both keyword matching rules and generative AI models to assign scores to candidates enhances the accuracy of job recommendation. Some of the other innovative capabilities of this platform include project-based learning, follow-up quiz , performance analytics, and remote proctoring.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial Intelligence, Career Recommendation System, Skill Gap Analysis, Personalized Learning, Generative AI, FastAPI, PostgreSQL, Job Recommendation, Resume Analysis, Adaptive Quiz System, Learning Roadmap Generation, Proctoring System, Web-Based Application

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

User behaviour analysis through opinion mining employs natural language processing to identify and quantify subjective information in textual data. This paper presents a systematic comparative evaluation of sentiment classification methods applied to product reviews from major e-commerce and food delivery platforms including Amazon, Flipkart, and Zomato. We evaluate five major methodological approaches: lexicon-based methods (VADER), classical machine learning models (TF-IDF with Naive Bayes and SVM), word embedding techniques (FastText and GloVe), deep learning architectures (CNN, LSTM, and BiLSTM), and state-of-the-art transformer models (BERT, DistilBERT, RoBERTa, ALBERT, and XLNet). Extensive experiments conducted on large-scale public datasets demonstrate that transformer-based models achieve the highest accuracy rates (92-96%), while FastText provides a highly competitive baseline (~93-95%). We conduct comprehensive qualitative error analysis using confusion matrices and employ explainability tools (LIME and SHAP) to understand model decision-making processes and identify failure patterns. Additionally, we address critical ethical considerations including data privacy protection, algorithmic bias detection, and fairness across demographic groups. Our findings establish a rigorous benchmark and provide a methodological blueprint for reproducible user behaviour analysis research, offering practitioners evidence-based guidance for selecting appropriate methods across different application contexts and performance requirements. Sent User behaviour analysis research.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Sentiment Analysis, Opinion Mining, Product Reviews, Transformer Models, BERT, RoBERTa, Explainability, Machine Learning, Natural Language Processing.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The increasing threat of explosive devices in public and sensitive environments has created a strong need for safer disposal techniques that minimize human involvement. This work presents the design and implementation of a teleoperated bomb disarming robotic arm, developed to perform hazardous tasks remotely while providing real-time visual feedback. The system is built around a Wi-Fi-enabled ESP32 microcontroller, integrated with a four-degree-of-freedom robotic arm and a mobile rover platform. Control is achieved through a web-based interface, allowing intuitive operation from a safe distance. Servo motors driven by a PWM controller enable precise manipulation of objects, while an ESP32-CAM module delivers live video streaming to the operator. Experimental results demonstrate reliable remote operation, reduced human exposure to danger, and effective handling of hazardous objects.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

ESP32 microcontroller, 4 DOF Robotic Arm, 4-wheel rover, ESP32 CAM module

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Work culture plays a significant role in enhancing employee engagement by influencing motivation, satisfaction, and organizational commitment. This study examines the impact of different types of work culture, including flexible, appreciative, collaborative, ethical, transformational, and developmental cultures, on employee engagement. A descriptive research design was adopted, and data were collected from 152 employees using a structured questionnaire. Statistical tools such as percentage analysis, correlation, regression, chi-square test, and SEM using SPSS were applied for analysis. The findings reveal that appreciative and flexible work cultures have the highest impact on employee engagement, followed by collaborative and ethical cultures. The study concludes that a positive work culture significantly enhances employee engagement and organizational performance.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Work Culture, Employee Engagement, Job Satisfaction, Organizational Performance, SPSS, SEM

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In an era where intelligent systems are rapidly evolving, This project presents the Design and Development of a Personal AI Autonomous system with multi agents . The AI Framework that integrates large language models with multi-agent orchestration to perform real-world tasks intelligently and efficiently. Unlike traditional chatbots, the proposed system is capable of understanding user intent, maintaining contextual awareness, The system is built using a modular architecture consisting of an orchestrator, multiple task-specific agents, and a retrieval-augmented generation (RAG) module. The orchestrator leverages advanced language models to interpret user queries, plan actions, and coordinate tool execution. Agents are responsible for handling specific tasks such as system control operations, email communication, scheduling reminders, and WhatsApp messaging. The integration of a RAG-based knowledge retrieval system enables the assistant to provide accurate, context-driven responses from external documents, To ensure personalization, the system incorporates encrypted user profiles, allowing the assistant to adapt responses based on user preferences and context securely . Additionally, the framework supports multiLLM integration.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

AI system with Multi Agents

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The building surfaces in Indian traditional architecture are not only considered mere enclosures but are treated as artistic and expressive mediums which can integrate functionality, aesthetics and cultural expression. This thesis paper intends to explore the craft of jali construction, mirror work (thikri) and Mewar miniature painting from Rajasthan, India. These crafts are part of traditional design vocabulary, developed out of the cultural heritage of the region and passed on from generation to generation. Each of these crafts plays a significant role in surface articulation and interior spatial composition. Jali, in addition to being decorative and ornamental is functional in many ways - it controls the temperature within the space by providing shade from the heat, regulates privacy of the space, while being beautiful at the same time. Similarly, mirror work (thikri) is traditionally used for decoration purposes, to enhance the beauty of the space by reflecting light within it, thus increasing its luminosity. Traditionally, miniature paintings had an important significance by portraying various narratives. With industrialization and changing aesthetic and functional preferences, the importance and meaning of these crafts has diminished and they are today used mostly as decorative elements. The aim of this thesis is to document and explore these crafts and their application within contemporary interiors in the context of their cultural heritage. Case study analysis of traditional interiors will be used in combination with visual and material analysis.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Jali, Thikri Work, Mewar Miniature Painting, Traditional Indian Crafts, Building Surface Crafts, Cultural heritage, Traditional craft techniques, Contemporary Applications

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The rapid adoption of electric vehicles has created an urgent need for efficient battery monitoring and management systems to ensure safety, longevity, and optimal performance. This paper presents the design and implementation of an Internet of Things (IoT) based electric vehicle battery monitoring system utilizing Arduino UNO microcontroller, ESP8266 Wi-Fi module, and various sensors including voltage divider, ACS712 current sensor, and LM35 temperature sensor. The proposed system enables real-time monitoring of critical battery parameters such as voltage, current, and temperature, with data visualization through both local LCD display and cloud based ThingSpeak platform. The integration of wireless communication facilitates remote monitoring capabilities, allowing users to track battery health from anywhere via internet connectivity. Experimental results demonstrate the system's ability to accurately measure battery parameters with a response time of less than 2 seconds and an accuracy of approximately 95%. The system provides early warning alerts for abnormal conditions, thereby enhancing EV safety and preventing potential battery failures. This cost effective solution offers a practical approach for smart energy management in electric vehicles and can be extended for integration with Battery Management Systems (BMS) for comprehensive battery health assessment.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

The project keywords are: IoT, EV Battery Monitoring, Arduino, ESP8266, Smart Energy Systems, ThingSpeak, Battery Management, Electric Vehicles, Wireless Sensor Networks.

  License

Creative Commons Attribution 4.0 and The Open Definition