Volume 13 | Issue 5 |

 Abstract

The growing urbanization and city population expansion have resulted in an immense rise in vehicle congestion, necessitating the need for advanced and responsive traffic control systems. The current research offers an advanced, real-time traffic management system implemented using Raspberry Pi and the YOLOv5 object detection engine. The system takes a real-time video feed as input and detects and classifies various vehicle classes, dynamically adjusting traffic signal timings based on vehicle density in particular lanes. One notable feature of the system is the detection of emergency vehicles--ambulances and fire trucks--and overriding regular signal functioning to give them constant right-of-way. The implementation is highly accurate, with a mean Average Precision (mAP@0.5) of 0.865, detection accuracy of 97% for ambulances, and a detection accuracy of 94% for fire trucks. Additionally, an easy-to-use web interface implemented using Flask allows real-time monitoring and control. This smart system significantly improves urban traffic flow and allows earlier emergency response, and it is a promising scalable part of smart city infrastructure.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial Intelligence, Computer Vision, Emergency Vehicle Detection, Intelligent Traffic System, Smart City, Traffic Management, YOLOv5

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Project Based Learning (PBL) is a teaching method that encourages students to solve real-world problems. This work focuses on project-based learning and culminates in Indian universities as teachers find ways to make learning meaningful and engaging for boys and girls. Project-based learning begins with a problem or task. Practical and realistic issues or problems that affect students' knowledge and development levels form the -basis of project-based learning teaching. Teachers teach students by asking questions, finding information, and looking at information. Encourages important skills such as teamwork, independent work, self-evaluation, time management, study, or oral and written communication. There are many benefits of using PBL in Indian universities. First, PBL can help students develop 21st-century skills such as problem-solving, critical thinking, and collaboration. These skills are essential to being successful at work and in life. Second, VET can help students learn better by engaging in active learning.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Instructional approach, group work, autonomous learning, self-assessment skills, time planning, problem-solving, critical thinking, collaboration, etc.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Sentiment analysis, also known as opinion mining, is a natural language processing (NLP) technique used to determine the emotional tone behind textual data. This project focuses on implementing sentiment analysis on product reviews to automatically classify customer feedback as positive, negative, or neutral. By leveraging machine learning algorithms and text processing techniques, the system analyzes user-generated content from e-commerce platforms to extract valuable insights into customer satisfaction. The goal is to assist businesses in understanding consumer sentiment at scale, enhancing product development, marketing strategies, and customer experience. Experimental results demonstrate that the model effectively captures sentiment polarity with high accuracy, showcasing its potential as a decision-support tool in online retail.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

SentimentAnalysis, OpinionMining, Product review analysis, Natural language processing(NLP),Customer Feedback analysis.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Indian democracy is facing the challenge of deepening its democratic form of governance at the local level but presently our civil society, intellectuals and political class do not realize it. The 73rd and 74th constitutional amendments which were initiated with the purpose of empowering the local governance but this democratic revolution has been hijacked by the ruling elites governing the different states with the help of district administrative setup. Panchayats and Municipal governments are not really empowered to legislate and administer upon the subjects devolved to them by the constitutional amendments because the state are not willing to part with their powers and vested interests.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Failure of Genuine Democratic Decentralization & Question of Substantive Democracy in India

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Setting up a secured web server environment on Arch Linux is a multifaceted process that entails configuring a lightweight and efficient server with robust security mechanisms. Arch Linux, known for its simplicity and flexibility, provides the ideal platform for deploying tailored web services. This guide details the essential components of creating a secure web server, focusing on software installation, configuration, access controls, and implementing best security practices. The tools involved include OpenSSL for TLS/SSL implementation, UFW or iptables for firewall configuration, and Fail2Ban for intrusion prevention. Existing systems often lack comprehensive security measures, leaving them vulnerable to various cyber threats. The proposed system focuses on creating a highly secure web server environment on Arch Linux by implementing multiple layers of security. This includes the latest encryption protocols, stringent firewall rules, and proactive intrusion prevention techniques, significantly enhancing the server's resilience against attacks.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Arch Linux, secured web server, Nginx, Apache, SSL/TLS, Let's Encrypt, firewall, UFW, iptables, intrusion detection systems (IDS)

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This project proposes a Neural Intrusion Detection System (Neural IDS) that leverages deep learning techniques to detect and classify cybersecurity threats in real time. By applying artificial neural networks to analyze network traffic patterns, the system achieves higher detection rates and reduced false positives compared to traditional signature-based systems. The model is trained on benchmark datasets, learning complex attack signatures and behavioral anomalies. The integration of intelligent threat detection into network security infrastructures ensures proactive defense, adaptability to novel threats, and automated response mechanisms..

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keywords-- Intrusion Detection System, Cybersecurity, Deep Learning, Neural Networks, Threat Detection, Anomaly Detection

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Bhartiya Siksha Me Swa - Viniyam Adhigam Ka Vikash : Vaidik Kal Se Adhunik Kal Tak

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Bhartiya Siksha Me Swa - Viniyam Adhigam Ka Vikash : Vaidik Kal Se Adhunik Kal Tak

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This comprehensive study within Kerala's educational framework investigates the complex interplay between social intelligence, academic effectiveness, and student performance among college professors. The research employed mixed methodologies to illustrate a significant and positive relationship between social intelligence and teaching effectiveness. It enhances the influence of student satisfaction on teaching efficacy and the well-being of educators. The association's sagacity is evidently manifest. Educators with advanced social intelligence excel in creating inclusive and engaging learning environments, significantly influencing student behaviour and promoting a harmonious classroom atmosphere. These findings highlight the crucial importance of social intelligence in formulating effective teaching methodologies and enhancing the overall educational experience in culturally diverse institutions in Kerala. Consequently, this study holds substantial importance for educator training and professional development. It aims to equip educators with the interpersonal and emotional intelligence necessary to enhance students' lives and promote their success in varied educational settings.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Social Intelligence, Job Performance Student Behaviour, Student Satisfaction.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In this paper, we describe the design of an Automated Content Platform which employs multi-agents systems for content collection, processing, and distribution. The described work merges domain knowledge of Large Language Models (LLMs) with practical deployment approaches that are bounded by scalability, modularity, and efficiency. Our approach followed a stepwise procedure starting with a web-scraping-based content acquisition, then proceeding to content summarization with several models including Gemini by Google, followed by distribution through Telegram and email. The components of the architecture are organized into streams, where each is an independent agent of the system, utilizing FastAPI and Streamlit interfaces, MongoDB for data storage, and so forth. Results prove that the platform substaially meets the requirements for efficient content acquisition from multiple sources, automated summarization of a wide range of topics, and prompt distribution through numerous channels. User testing with 25 participants indicated 87% satisfaction with the quality of summaries provided and 92% with usability of the platform. The measured response times were 3.2 seconds to retrieve content and 4.5 seconds to provide a summary of the retrieved content. The work provided in this document adds to the existing body of knowledge by showing that automated content processes can be further enhanced by multi-agent frameworks focusing on system agility and flexibility in operations.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Content Automation, Multi-Agent Systems, Natural Language Processing, Web Scraping, Content Distribution, Large Language Models, Retrieval-Augmented Generation

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Medical datasets often have a skewed class distribution and a lack of high-quality annotated images. However, deep learning methods require a large amount of labelled data for classification. In this study, we present a few-shot learning approach for the classification of ultrasound breast cancer images using meta-learning methods. We used prototypical networks and model-agnostic meta-learning (MAML) algorithms as meta-learning methods. The breast ultrasound images (BUSI) dataset, which has three classes and is difficult to use in meta-learning, was used for meta-testing in a cross-domain approach along with other datasets for meta-training. Our proposed approach yielded an accuracy range of 0.882-0.889, achieved by implementing the ResNet50 backbone with ProtoNet in a 10-shot setting. These results represent a significant improvement ranging from 6.27 to 7.10% over the baseline accuracy of 0.831. The results showed that ProtoNet outperformed the MAML method for all k-shot settings. In addition, the use of ResNet models as the backbone network for feature extraction was found to be more successful than the use of a four-layer convolutional model. Our proposed method is the first attempt to apply meta-learning for few-shot classification in the BUSI dataset while providing higher accuracy compared to deep learning methods for medical images with small-scale datasets and few classes. The methodology used in this study can be adapted to other datasets with similar problems.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Few-shot classification, meta learning, MAML, breast cancer.

  License

Creative Commons Attribution 4.0 and The Open Definition