Volume 13 | Issue 3 |

 Abstract

The exponential growth of digital image data necessitates advanced systems for efficient retrieval based on visual content. This project proposes a Content-Based Image Retrieval (CBIR) system that leverages machine learning techniques to bridge the gap between low-level image features and high-level user intent. The system extracts color histograms and deep learning-based features using pre-trained models like ResNet50 and VGG16 , combining them into hybrid feature vectors for enhanced representation. To ensure scalability, FAISS (Facebook AI Similarity Search) is employed for indexing and rapid similarity comparisons using metrics like cosine similarity and Euclidean distance . Machine learning classifiers such as SVM (80% accuracy) and KNN (60% accuracy) are trained on color histogram features, while a fine-tuned ResNet50 model (97.5% accuracy) enhances deep feature extraction. The system processes user queries in real time, delivering results through an intuitive interface and addressing challenges like computational efficiency and semantic understanding. Evaluated on datasets like Corel 1K, Vistex, and Faces, the system demonstrates superior performance in retrieving visually similar images, with applications in e-commerce, medical imaging, and forensic analysis. By integrating traditional and modern techniques, this CBIR system sets a benchmark for accuracy, scalability, and user-centric design in image retrieval.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Content-Based Image Retrieval (CBIR), ResNet50, FAISS, SVM, KNN, color histograms, feature extraction, cosine similarity, Euclidean distance, hybrid features, real-time retrieval.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The AI-powered YouTube Transcript Summarizer provides a new way to consume video content more efficiently through summarized, condensed versions of YouTube videos that are really long with just a simple Chrome Extension. It takes advantage of advanced NLP techniques together with the Hugging Face Transformers library to extract meaningful insights from the transcripts of videos, whereby users will be able to get the vital pieces of information without having to actually watch complete videos. It combines three basic key components: a Text Summarizer summarizing important information, Hugging Face Transformers to ensure coherence and preciseness of summaries, and a Python-based Web API that communicates the extension with the back-end services seamlessly. A comparative study highlights that such a tool serves in enhancing the user experience of decreasing exposure to irrelevant or misleading information and provides substantive improvements to educational It is designed for professional and personal uses of videos. This tool will be going to change the way users consume online video content. The present fast-moving scenario of today's world faces problems of video consumption, which is a time-consuming and inefficient process.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

AI-powered YouTube Transcript Summarizer, Video Content Efficiency, Chrome Extension, Natural Language Processing (NLP)

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

With the increasing prevalence of online fraud and the use of fake logos to deceive consumers, there is a need for effective methods to detect and prevent the use of fake logos on the internet. In this paper, we propose a method for detecting fake logos using machine learning techniques. Our approach involves extracting features from the logos and training a classifier to distinguish between real and fake logos. We evaluate the performance of our method on a dataset of real and fake logos and demonstrate its effectiveness in detecting fake logos with high accuracy. Every day, hundreds of domain names, websites and logos are being cloned by cyber criminals who want to gain your trust so they can steal your data. It is becoming a big issue in the online world and needs to be addressed. This article will discuss the initial project background of our new Online Fake Logo Detection System.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Deep learning, Convolutional Neural Networks (CNNs), Image recognition Brand protection, Counterfeit detection, Image processing.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

A nation's economic and social structure depends on its capital market.It serves the vital functions of raising capital for the public and private sectors, fostering stability and balance in the financial system, reducing reliance on the banking industry, advancing the economy, generating employment, and providing an alternate means of conserving money.A robust capital market will mitigate the effects of economic volatility, which may be exacerbated by the rapid flow of money. The Thai capital market is still plagued by a number of problems, nevertheless, including a small number of institutional investors, a small number of retail investors, a lack of effective regulatory enforcement, a restricted selection of financial products, and high transaction costs.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Capital market

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Performance appraisals are essential for evaluating employee contributions, coordinating business objectives, and promoting professional development. Employee perceptions of assessment systems are examined in this study, with a focus on their efficacy, equality, and transparency as well as their effects on career advancement, work satisfaction, and motivation. Using survey data and secondary research, the study investigates the connection between organizational commitment, productivity, and performance reviews. Results indicate that workers like open assessments that acknowledge accomplishments and offer helpful criticism. But issues like subjectivity and uneven grading draw attention to how difficult it is to maintain justice. The paper highlights the necessity of objective requirements for performance, standardized tools, and continuous feedback mechanisms. Engagement, motivation, and trust are all increased by a well-organized assessment process, which eventually boosts employee morale and company performance.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Performance Appraisals, Employee Perceptions, Career Advancement, Continuous Feedback, Organizational Commitment

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In today's rapidly evolving global markets and shifting consumer preferences, organizations face a broad spectrum of sustainability challenges that go beyond the traditional Environmental, Social, and Governance (ESG) frameworks. While ESG metrics continue to play a pivotal role, the complexity of sustainability now encompasses a wider array of economic, social, and environmental considerations. These emerging challenges include the integration of circular economy principles, responsible innovation, the impact of digital transformation on resource consumption, evolving regulatory landscapes, and the demand for greater transparency and accountability across supply chains. As organizations navigate these pressures, they must adopt a more holistic approach to sustainability--one that transcends conventional ESG paradigms and fosters long-term value creation. This paper explores these multifaceted sustainability challenges and provides a comprehensive analysis of the strategies and frameworks that organizations can leverage to maintain resilience, promote sustainable growth, and respond to the demands of an increasingly conscientious global market.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Sustainability, Stake Holders, Supply Chain, Demand Chain, Integration, Innovation, Value Maximization

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Automated inspection systems that utilize image processing techniques play a vital role in industries requiring high precision, efficiency, and quality control. This paper explores the design, development, and implementation of these systems, which rely on high-resolution imaging devices and advanced algorithms to evaluate product quality and identify defects. By replacing manual inspection, these technologies boost productivity while minimizing human error and operational costs. Techniques such as edge detection, pattern recognition, feature extraction, and machine learning enable automatic defect identification and classification. Real-time analysis combined with automated decision-making ensures continuous and accurate inspection, benefiting industries like automotive, electronics, pharmaceuticals, and food processing. Despite their advantages, challenges such as variable lighting conditions, occlusions, and the need for large datasets to train machine learning models remain. The incorporation of deep learning and AI-driven optimizations strengthens defect detection and system performance. Ultimately, these systems enhance industrial automation, driving improvements in efficiency, accuracy, and cost-effectiveness while maintaining stringent quality standards.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Automatic inspection system, Image processing, Edge detection, Quality control, Defect Recognition

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Abstract In this paper, we present a machine learning models for handwritten Kannada language district name recognition. Kannada, is a prominent Indian language, has a unique script with complex characters, making handwritten recognition a challenging task. The goal of this research is to develop an efficient system that can accurately recognize district names in Kannada from handwritten text. The proposed method involves preprocessing of handwritten input images, feature extraction using deep learning techniques, and classification using machine learning algorithms such as Convolutional Neural Networks (CNNs). A dataset containing a large variety of handwritten Kannada district names is utilized for training and testing the model. We evaluate the performance of various machine learning models, comparing their accuracy and processing speed. The results demonstrate the effectiveness of deep learning models in achieving high recognition accuracy, even with diverse and noisy handwritten inputs. This system holds potential for applications in digitizing official documents, automating postal services, and improving language-based AI systems for Kannada. The study also explores future directions for improving recognition accuracy, including the integration of hybrid models and dataset expansion.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords Handwritten Text Recognition, Kannada Language, Machine Learning, Convolutional Neural Networks (CNN), District Name Recognition, Feature Extraction, Deep Learning, Optical Character Recognition (OCR), Image Processing, Artificial Intelligence (AI), Pattern Recognition, Language Digitization, Handwritten Kannada Dataset.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The rise of human civilization is an awe inspiring saga that is equally remarkable and intriguing. The growth from being a caveman to becoming the rocket scientist is an enigmatic success story. However when it comes to attributing the credit of this accomplishment, gender plays the decisive role. Man basks in the glory of achievements while woman is pushed ruthlessly to the periphery. Over the past few hundred years, literature has sincerely produced such priceless narratives that have explicitly documented the plight of marginalized, particularly women. Shashi Deshpande is one of the contemporary writers in English literature whose oeuvre explores the inner lives of suffering women. Deshpande's protagonists are marginalized women who struggle for their autonomy, expression and belonging. In addition, there are women in her works who are so acutely suppressed that they do not make any effort to venture out of the position assigned to them by patriarchy. They internalise the intent of subjugating forces and remain at the margin forever. The present paper explores these women who have lost all hopes of crossing the threshold and seem to have willingly succumbed to their situations. The writer's focus, albeit small, on these women indicates that women liberation is still a far cry. Besides, the suffering embitters these marginalized women to such an extent that they too become agents of oppressive patriarchal norms. They are sufferers as well as the perpetrators imposing misery on other women.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

marginalized, oppressed, subjugated, suffering, periphery, suffering, patriarchy

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Through this research paper an attempt has been made to highlight the changes in education from ancient times to the present.Whereas in ancient times, through education, man wanted to achieve worldly and spiritual goals along with economic goals in his life, in the present times, the aim of education is only to get a job and earn a livelihood,but it is not only the fault of the student but also of his family and the government who instead of making them aware about those goals emphasize only on accumulating money.Many foreign accounts reveal the high moral ideals of Indians, but in today's times, man looks only after his own interests due to which his moral degradation is taking place. After fourth or fifth class, no subject like moral education is seen in the school curriculum. Yes, these subjects have been added at the graduate level in NEP, the effect of which will definitely be seen in the next few years.In ancient times, education used to focus on all-round development, but as a result of the policies adopted by the British government in modern times, the aim of education in present India is only to get a job, meeting the western standards etc. are left, we are adopting the education system created by the British government during the colonial period in some form or the other till date and from there the downfall of our education, moral values and society started.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Education, Vedas, All-round Development, Madrasa, School

  License

Creative Commons Attribution 4.0 and The Open Definition