Air pollution, a silent public health crisis, demands smarter solutions. This research explores a novel approach: using machine learning and the Internet of Things (IoT) to predict hyper-local Air Quality Index (AQI). We integrate a Random Forest algorithm with NodeMCU for pinpoint-accurate AQI forecasts. The core lies in a robust machine learning model trained on historical data rich in environmental parameters-air quality indicators like PM2.5 and ozone, coupled with meteorological factors. This empowers the model to predict future AQI levels with exceptional accuracy. Real-time data acquisition is handled by NodeMCU, an IoT platform acting as our environmental guardian. It gathers live data and feeds it to the model, enabling real-time, location-specific AQI predictions. The model's effectiveness is rigorously evaluated, achieving a promising 99% accuracy-a testament to its potential to revolutionize air quality management. By harnessing the power of machine learning and IoT, this research paves the way for a future where proactive air quality management fosters healthier cities and a more sustainable future.

In today's era websites have emerged as a means for businesses and organizations to connect with their customers. To ensure a user experience, across devices such as desktops, laptops, smartphones and tablets it is essential for a website to be both responsive and visually appealing. This survey paper aims to provide an overview of the methods recommended approaches and current patterns in designing and developing websites that're both responsive and visually captivating. The objective of this paper is to equip web developers, designers and researchers, with an understanding of the process involved in creating websites that're both responsive and visually appealing. As digital technologies continue to advance it is becoming increasingly vital to create engaging web experiences that cater to the user's needs.

Climatic changes, which are increasingly responsible for the increase in the incidence and intensity of natural calamities, demand prompt and precise response and recovery actions. This research aims at determining how AI computer vision methodologies can be applied to aerial and satellite images related to a disastrous management system. We study different AI techniques that can be used in damage assessment, resource allocation, and recovery planning. This paper reviews data sources starting with aerial imagery from drones, to aircraft and satellite imagery, and encompasses all the diversity in sensors and resolutions. Key AI techniques brought into discussion include CNNs, semantic segmentation, and logistics optimization models. Key challenges identified are data quality, model generalization, and ethics. Applications in real-world settings are illustrated through case studies including rapid damage assessment following hurricane Maria, monitoring the Australian bushfires, and optimization of resources allocation during the COVID-19 pandemic. The survey outlines proposed future research directions with an emphasis on real-time analysis, integration of IoT and edge computing, improvement of model interpretability, and collaborative platforms. Surveying creates demands for continuous innovation and collaboration and underlines improvement for enhanced disaster preparedness and resilience.

In today's era of digitalization, streamlining traffic management and ensuring public safety are paramount. Traditional methods of driver license verification often face challenges in terms of efficiency, accuracy, and real-time capabilities. To address these limitations, we propose a novel Driving License Verification System (DLVS) that leverages advanced facial recognition and optical flow techniques to revolutionize driver license verification. The DLVS seamlessly integrates with traffic management systems, enabling real-time verification of driver licenses, enhancing traffic communication, and promoting overall public safety.

Interactive art often struggles to dynamically reflect audience emotions in real-time. To address this, we developed Expressive Canvas, a system that integrates emotion detection with real-time image manipulation. This innovation aims to transform audience emotions into dynamic visual art, creating immersive experiences. Our approach employs a pose-to-emotion detection model using various mathematical equations to enhance accuracy. Trained on the FER-2013 dataset, our model achieved a test accuracy of 62.47%, despite significant challenges arising from class imbalances, particularly for the "Disgust" emotion. The system performs efficiently, detecting emotions in an average of 8.43 seconds and manipulating images in just 0.02 seconds. Field tests of Expressive Canvas across various events demonstrated its consistent performance and reliability. Despite existing challenges, the system successfully transformed audience emotions into visual expressions, enhancing the connection between art and viewers. This study highlights the potential for real-time emotional expression in interactive art. Future work will focus on mitigating class imbalances and optimizing the neural network to improve accuracy and expressiveness. Key terms include emotion detection, pose-to-emotion detection, real-time image manipulation, interactive art, and neural networks.

This study tackles the limitations of traditional distance metrics in machine learning, particularly their inadequacy in managing missing data and varying feature importance. We introduce a novel custom distance metric that offers a more nuanced and robust measure of similarity, addressing the complexities and imperfections of real-world data. Our extensive evaluation of the CustomKNN algorithm, leveraging this custom distance metric, reveals exceptional performance and robustness across diverse datasets, consistently outperforming traditional metrics in various classification tasks. The custom distance metric shows superior robustness to missing data and scalability, despite its higher computational complexity and memory requirements. Our findings highlight the versatility and effectiveness of the CustomKNN algorithm, establishing it as a formidable option for classification tasks. This research advances the development of more adaptable and context-aware distance metrics, significantly enhancing the accuracy and reliability of machine learning models in practical applications.