Object detection plays a crucial role in various applications, including surveillance, autonomous driving, and industrial automation, where accurate and timely identification of objects is essential. This research proposes a novel framework that combines the YOLOv8 backbone network with an attention mechanism and a Transformer-based detection head, significantly enhancing object detection performance in real-time images and video. The incorporation of attention mechanisms refines feature extraction from complex scenes, enabling the model to focus on relevant regions within images. Using the integration of Transformer architecture, the model leverages long-range dependencies and global context, leading to more accurate bounding box predictions. The proposed system effectively processes real-time data, demonstrating superior classification performance with precision rates reaching 96.78 % and recall rates of 96.89 %. The mean average precision (mAP) is calculated at 89.67 %, showcasing the framework's robustness across various practical scenarios. The framework is developed to address challenges in object detection, such as detecting multiple objects in crowded environments and varying lighting conditions. The Python architecture supports the implementation of the proposed model. The Python architecture supports the implementation of the proposed model. The results section assesses the Attention Transformer-YOLOv8 model against established algorithms like Faster R-CNN, YOLOv3, YOLOv5n, and SSD, utilizing metrics.

crime investigation in Cloud Computing Environments (CCE) includes meticulously examining digital evidence within cloud infrastructures in order to recognize and reduce cyber risks and criminal activity. The rising reliance on cloud technology exposes enterprises to advanced cybercrime, necessitating the need for and significance of forensic investigation in CCE. Existing approaches to forensic crime investigation frequently encounter scalability, efficiency, and adaptability issues due to the dynamic nature of cloud infrastructures. These constraints impede reliable and timely detection of cyber threats, stressing the need for novel techniques. To overcome these issues, this study provides a unique approach for forensic evidence recognition and classification in CCE using a hybrid Recurrent Neural Network (RNN) and Extreme Learning Machine (ELM). The methodology includes preprocessing based on Z-Score Normalization, data collecting, and cloud forensics evidence detection. A hybrid RNN-ELM model is put into practice, specifically designed for sequence modelling in cloud-based cybercrime data. By optimizing feature selection and boosting overall efficiency, Grey Wolf Optimization (GWO) helps to even more enhance the model's performance. The practical usefulness of the proposed approach was demonstrated by the implementation of the study's results in Python software. The proposed RNN-ELM method exhibits an average accuracy increase of 3.18% compared to existing methods, surpassing Deep Neural Network-Shuffled Frog Leap Optimization (DNN-SFLO) and Deep Learning Modified Neural Network (DLMNN) with accuracy percentages of 99.4%, 99.09%, and 96.25%, respectively. The created model offers a viable option for tackling the changing issues in cybercrime investigations within cloud environments. It demonstrates improved scalability, efficiency, and precision in managing forensic evidence within cloud computing scenarios.

In an era marked by a proliferation of online reviews across various domains, navigating the extensive and diverse range of opinions can be challenging. Sentiment analysis aims to extract and interpret sentiments from these vast pools of data using computational linguistics and information retrieval techniques. This study focuses on employing deep learning methods such as Deep Belief Networks (DBN) and Gated Recurrent Units (GRU) to classify reviews into positive and negative sentiments, addressing the issue of information overload in Product Reviews. The primary objective is to develop an efficient sentiment analysis system that reliably categorizes reviews as positive or negative. The study introduces a novel sentiment analysis framework combining Deep Belief Networks and Gated Recurrent Units for online product review classification, enhancing accuracy through advanced feature extraction and classification techniques. The comprehensive preparation pipeline—comprising data splitting, stemming, stop word removal and special character separation—enhances dataset refinement for improved classification accuracy. The proposed framework consists of four main phases: pre-processing, feature extraction, classification, and evaluation. During the preparation phase, the dataset is meticulously cleaned and refined to reduce noise and enhance signal quality. Significant features are then extracted from the pre-processed data using advanced feature extraction algorithms. The DBN-GRU model leverages these features for sentiment classification, effectively distinguishing between positive and negative attitudes. The framework's performance is subsequently evaluated to assess its efficacy in accurately classifying reviews. The combination of in-depth pre-processing procedures and the DBN-GRU technique yielded promising results in sentiment categorization. The framework demonstrated a high accuracy of 98.74% in differentiating between positive and negative sentiments, thereby facilitating the effective analysis of online reviews. This study presents a robust framework for sentiment analysis, utilizing the DBN-GRU method to classify online reviews. Through extensive preprocessing and advanced classification techniques, the system addresses the challenges of noise and information overload in online reviews, providing valuable insights for both consumers and businesses.

Early illness diagnosis, treatment monitoring, and healthcare administration all depend heavily on the identification of abnormalities in medical data. This paper proposes a unique way to improve healthcare anomaly detection through the integration of attention mechanisms and Generative Adversarial Networks (GANs) for improved performance. By integrating GANs, artificial data that closely mimics the distributions of actual healthcare data may be produced, so, it is important to supplementing the dataset and strengthening the resilience of anomaly detection algorithms. Simultaneously, the Convolutional Block Attention Module (CBAM) facilitates the model's concentration on useful characteristics present in the data, thereby augmenting its capacity to identify minute deviations from the norm. The suggested method is assessed using a large dataset from healthcare settings that includes both typical and unusual cases. When compared to current techniques, the results show notable gains in anomaly detection performance. The model also shows resilience to noise, small abnormalities, and class imbalance, indicating its potential for practical clinical applications. The suggested strategy has the potential to improve clinical decision-making and patient care by giving doctors faster, more precise insights into anomalous health states. With an accuracy of around 99.12%, the suggested GAN-CBAM is implemented in Python software and outperforms other current techniques such as Gaussian Distribution Anomaly detection (GDA), Augmented Time Regularized (ATR-GAN), and Convolutional Long Short-Term Memory (ConvLSTM) by 2.97%. With potential benefits for bettering patient outcomes and the effectiveness of the healthcare system, the suggested strategy is a major step forward in the improvement of anomaly identification in the field of medicine.

To facilitate smooth human-computer interaction (HCI) in a variety of contexts, from augmented reality to sign language translation, real-time gesture detection is essential. In this paper, researchers leverage federated convolutional neural networks (CNNs) to present a novel strategy that tackles these issues. By utilizing federated learning, authors may cooperatively train a global CNN model on several decentralized devices without sharing raw data, protecting user privacy. Using this concept, researchers create a federated CNN architecture designed for real-time applications including gesture recognition. This federated approach enables continuous model refinement and adaption to various user behaviours and environmental situations by pooling local model updates from edge devices. This paper suggests improvements to the federated learning system to maximize responsiveness and speed. To lessen the probability of privacy violations when aggregating models, this research uses techniques like differential privacy. Additionally, to reduce communication overhead and quicken convergence, To incorporate adaptive learning rate scheduling and model compression techniques research show how federated CNN approach may achieve state-of-the-art performance in real-time gesture detection tasks through comprehensive tests on benchmark datasets. In addition to performing better than centralized learning techniques. This approach guarantees improved responsiveness and adaptability to dynamic contexts. Furthermore, federated learning's decentralized architecture protects user confidentiality and data security, which qualifies it for usage in delicate HCI applications. All things considered, the design to propose a viable path forward for real-time gesture detection system advancement, facilitating more organic and intuitive computer-human interactions while preserving user privacy and data integrity. The proposed federated CNN approach achieves a prediction accuracy in real-time gesture detection tasks, outperforming centralized learning techniques while preserving user privacy and data integrity. The proposed framework that achieves prediction accuracy of 98.70% was implemented in python.

Traumatic Brain Injury (TBI) is a significant global health concern, often leading to long-term disabilities and cognitive impairments. Accurate and timely diagnosis of TBI is crucial for effective treatment and management. In this paper, we propose a novel federated convolutional neural network (FedCNN) framework for predictive analysis of TBI in decentralized health monitoring. The framework is implemented in Python, leveraging three diverse datasets: CQ500, RSNA, and CENTER-TBI, each containing annotated brain CT images associated with TBI. The methodology encompasses data preprocessing, feature extraction using gray level co-occurrence matrix (GLCM), feature selection employing the Grasshopper Optimization Algorithm (GOA), and classification using FedCNN. Our approach achieves superior performance compared to existing methods such as DANN, RF and DT, and LSTM, with an accuracy of 99.2%, surpassing other approaches by 1.6%. The FedCNN framework offers decentralized privacy-preserving training across individual networks while sharing model parameters with a central server, ensuring data privacy and decentralization in health monitoring. Evaluation metrics including accuracy, precision, recall, and F1-score demonstrate the effectiveness of our approach in accurately classifying normal and abnormal brain CT images associated with TBI. The ROC analysis further validates the discriminative ability of the FedCNN framework, highlighting its potential as an advanced tool for TBI diagnosis. Our study contributes to the field of decentralized health monitoring by providing a reliable and efficient approach for TBI management, offering significant advancements in patient care and healthcare management. Future research could explore extending the FedCNN framework to incorporate additional modalities and datasets, as well as integrating advanced deep learning architectures and optimization algorithms to further improve performance and scalability in healthcare applications.

Employing deep learning techniques on fMRI data enables the exploration of universal and culturally specific neural correlates underlying language processing across diverse populations. The study presents "BrainLang DL," a novel deep learning (DL) approach leveraging functional Magnetic Resonance Imaging (fMRI) data to unveil neural correlates of language processing across diverse cultural backgrounds. To bridge the knowledge gap in the universal and culture-specific aspects of language processing, we engaged participants from various cultural groups in a series of linguistic tasks while recording their brain activity using fMRI. Our rigorous data preprocessing pipeline included steps such as motion correction, slice timing correction, and spatial smoothing to enhance data quality for subsequent analysis. For feature extraction, research utilized the Crocodile Hunting Optimization (CHO) algorithm to pinpoint critical brain regions and connectivity patterns linked to language functions. To capture the temporal dynamics of neural activity related to language processing, we deployed advanced recurrent neural networks, specifically Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) models. These techniques enabled us to unravel how linguistic information is encoded and processed over time. Our findings reveal both common and unique neural activation patterns in language processing across different cultures. Universally shared neural mechanisms highlight the fundamental aspects of language processing, while distinct variations underscore the influence of cultural context on brain activity. Furthermore, we employed Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks to analyze the temporal dynamics of language-related neural activity, uncovering how linguistic information is represented and processed over time. By integrating DL with fMRI analysis, our study provides a nuanced understanding of the neural correlates of language across cultures. It reveal both shared neural mechanisms underlying language processing across diverse populations and culturally specific variations in brain activation patterns. These findings contribute to a more comprehensive understanding of the neural basis of language and its modulation by cultural factors. Ultimately, our approach offers insights into the complex interplay between language, cognition, and culture, with implications for fields such as linguistics, neuroscience, and cross-cultural psychology.

Technological improvements have led to a rapid expansion of the digital realm, raising concerns about cyber security. The last ten years have seen an enormous rise in Internet applications, which has greatly raised the requirement for information network security. In the realm of cyber security, detecting anomalies efficiently and effectively is paramount to safeguarding digital assets and infrastructure. Traditional anomaly detection methods often struggle with the evolving landscape of cyber threats, particularly in distributed environments. To address this challenge, the research proposes a novel approach leveraging federated learning and Long Short-Term Memory (LSTM) networks. Federated learning permits training models across decentralised data sources without sacrificing data privacy, and LSTM networks are highly effective in identifying temporal correlations in sequential data, which makes them suitable for analysing cyber security time-series data. In this paper, the study presents the federated LSTM model architecture tailored for anomaly detection in distributed environments. By allowing model updates to be performed locally on individual devices or servers without sharing raw data, federated learning mitigates privacy concerns associated with centralized data aggregation. This decentralized approach not only safeguards sensitive information but also fosters collaboration among diverse stakeholders, empowering them to contribute to model improvement without relinquishing control over their data. Python software is used to implement the method. The research demonstrate its effectiveness through experiments on real-world cyber security datasets, showcasing improved detection rates compared to traditional methods. When compared to RNN, SVM, and CNN, the suggested Fed LSTM method exhibits superior accuracy with 98.9%, which is 2.28% more advanced. Additionally, the research discuss the practical implications and scalability of our approach, highlighting its potential to enhance cyber security measures in distributed threat scenarios.

Diabetic retinopathy (DR) is a severe complication of diabetes mellitus, leading to vision impairment or even blindness if not diagnosed and treated early. A manual inspection of the patient's retina is the conventional way for diagnosing diabetic retinopathy. This study offers a novel method for the identification of diabetic retinopathy in medical diagnosis. Using a hybrid Generative Adversarial Network (GAN) and Bidirectional Gated Recurrent Unit (BiGRU) model, further refined using the African Buffalo Optimization algorithm, the model's capacity to identify minute patterns suggestive of diabetic retinopathy is improved by the GAN's skill in extracting complex characteristics from retinal pictures. The technique of feature extraction plays a critical role in revealing information that may be hidden yet is essential for a precise diagnosis. Then, the BiGRU part works on the characteristics that have been extracted, efficiently maintaining temporal relationships, and enabling thorough information absorption. The combination of GAN's feature extraction capabilities with BiGRU's sequential information processing capability creates a synergistic interaction that gives the model a comprehensive grasp of retinal pictures. Moreover, the African Buffalo Optimization technique is utilized to optimize the model's performance for improved accuracy in the identification of diabetic retinopathy by fine-tuning its parameters. The current study, which uses Python, obtains a 98.5% accuracy rate and demonstrates its amazing ability to reach high levels of accuracy in Diabetic Retinopathy Detection.

The rapidly developing field of "Commercial Operation Divergence Analysis," this research seeks to identify and understand differences in commercial systems that exceed expected results. Approaches in this domain aim to identify the characteristics of process implementations that are associated with changes in process effectiveness. This entails identifying the features of procedural behaviours that result in unpleasant results and figuring out which behaviours have the biggest impact on increased efficiency. As the scale and complexity of big data management and process mining continue to expand, the threat of cyber-attacks poses a critical challenge. This research leverages machine learning techniques for the detection and defence against cyber threats within the realm of big data management and process mining. The study introduces novel metrics such as Skewness, Coefficient of Variation, Standard Deviation, Maximum, Minimum, and Mean for assessing the security state, utilizing variables like SPI, SPEI, and SSI. The research addresses prior issues in cyber-attack detection by integrating machine learning into the specific context of big data and process mining. The novelty lies in the application of Skewness and other statistical metrics to enhance the precision of threat detection. The results demonstrate the effectiveness of the proposed methodology, showcasing promising outcomes in identifying and mitigating cyber threats in the given dataset and which makes use of Support Vector Regression (SVR), has a standard deviation of 0.9, which is consistent with the variability shown in SVM. The results demonstrate a significant achievement, with a Mean Absolute Error (MAE) of 0.98, indicating the efficacy of the proposed approach in providing accurate and timely insights for cyberattack detection and defense, thereby enhancing the overall security posture in data-intensive systems. The results highlight how well the proposed method extracts significant insights from complicated event data, with important ramifications for real-world application and decision-making procedures.

Advancements in data capture techniques in the field of Magnetic Resonance Imaging (MRI) offer faster retrieval of critical medical imagery. Even with these advances, reconstruction techniques are generally slow and visually poor, making it difficult to include compression sensors. To address these issues, this work proposes a novel hybrid GAN-DRN architecture-based method for MRI reconstruction. This approach greatly improves texture, boundary characteristics, and picture fidelity over previous methods by combining Generative Adversarial Networks (GANs) with Deep Residual Networks (DRNs). One important innovation is the GAN's all-encompassing learning mechanism, which modifies the generator's behaviour to protect the network against corrupted input. In addition, the discriminator assesses forecast validity thoroughly at the same time. With this special technique, intrinsic features in the original photo are skillfully extracted and managed, producing excellent results that adhere to predetermined quality criteria. The Hybrid GAN-DRN technique's effectiveness is demonstrated by experimental findings, which use Python software to achieve an astounding 0.99 SSIM (Structural Similarities Index) and an amazing 50.3 peak signal-to-noise ratio. This achievement is a significant advancement in MRI reconstructions and has the potential to completely transform the medical imaging industry. In the future, efforts will be directed towards improving real-time MRI reconstruction, going multi-modal MRI fusion, confirming clinical effectiveness via trials, and investigating robustness, intuitive interfaces, transferable learning, and explanatory techniques to improve clinical interpretive practices and adoption.

The combination of blockchain technology and smart contracts has become a viable way to expedite claims processing and payouts in the quickly changing insurance industry. Enhancing efficiency, transparency, and reliability for the industry may be achieved by automating certain procedures and initiating them on predetermined triggers, smart contracts that is event-based. Conventional insurance procedures can be laborious, slow, and prone to human mistake, which can cause inefficiencies and delays in the resolution of claims. This research proposes a simplified system that automates the whole claims process from submission to reimbursement by utilizing blockchain technology and smart contracts. The suggested method does away with the requirement for human claim filing by having policyholders' claims automatically triggered by predetermined occurrences. These occurrences might be anything from medical emergencies to natural calamities, enabling prompt and precise claim start. The whole claims process is managed by smart contracts that are programmed with precise triggers and conditions, guaranteeing transaction immutability, security, and transparency. Moreover, reimbursements are carried out automatically after the triggering event has been verified, disregarding conventional bureaucratic processes and drastically cutting down on processing times. This strategy decreases the possibility of fraud and disagreement while also improving operational efficiency by combining self-executing contracts with decentralized ledger technology. Insurance companies and policyholders will both eventually profit from an accelerated, transparent, and reliable claims processing procedure thanks to the use of event-based smart contracts. A Python-implemented system achieving 97.6% accuracy using the proposed method, demonstrates its efficacy and reliability for the given task.

Forecasting water quality is critical to environmental management because it facilitates quick decision-making and resource allocation. On the opposite hand, current methods are not always able to produce reliable forecasts, which is often due to challenges in parameter optimization for complex models. This research presents a novel approach to enhance the forecasting accuracy of water quality by optimizing neuro-fuzzy models using Tunicate Swarm Optimisation (TSO). The introduction highlights the limitations of current techniques as well as the necessity for precise estimates of water quality. One of the drawbacks is that neuro-fuzzy models are not well-modelled, which makes it harder for them to identify the minute patterns in data on water quality. The suggested approach is unique in that it applies TSO, an optimization algorithm inspired by nature that emulates tunicates' behaviour, to the neuro-fuzzy models' parameter optimization process. The highly complex parameter space is effectively navigated by TSO's swarm intelligence, which strikes a balance between exploration and exploitation to improve model performance. To optimize model parameters, the process comprises three steps: creating an objective function, defining the neuro-fuzzy model, and seamlessly integrating TSO. By mimicking the motions of tunicates as they look for the best conditions in the marine environment, TSO constantly optimizes the variables. Experiments demonstrate that the proposed strategy is more effective than traditional optimization techniques in forecasting water quality. As seen by the optimised neuro-fuzzy model's increased prediction accuracy and several dataset validations, Tunicate Swarm Optimisation has potential for reliable environmental forecasting. This work presents a potential path for improved environmental decision-making systems by offering an optimisation strategy inspired by nature that overcomes the limitations of existing methods and enhances water quality forecasting tools.

The goal of the present research is to better understand the need of accurate and ongoing monitoring in the complicated chronic metabolic disease known as diabetes. With the integration of an intelligent system utilising a hybrid adaptive machine learning classifier, the suggested method presents a novel way to tracking individuals with diabetes. The system uses cutting edge technologies like intelligent tracking and machine learning (ML) to improve the efficacy and accuracy of diabetes patient monitoring. Integrating smart gadgets, sensors, and telephones in key locations to gather full body dimension data that is essential for diabetic health forms the architectural basis. Using a dataset that includes comprehensive data on the patient's characteristics and glucose levels, this investigation looks at sixty-two diabetic patients who were followed up on a daily basis for sixty-seven days. The study presents a hybrid architecture that combines a Convolutional Neural Network (CNN) with a Support Vector Machine (SVM) in order to optimise system performance. To train and optimise the hybrid model, Grey Wolf Optimisation (GWO) is utilised, drawing inspiration from collaborative optimisation in wolf packs. Thorough assessment, utilising standardised performance criteria including recall, F1-Score, accuracy, precision, and the Receiver Operating Characteristic (ROC) Curve, methodically verifies the suggested solution. The results reveal a remarkable 99.6% accuracy rate, which shows a considerable increase throughout training epochs. The CNN-SVM hybrid model achieves a classification accuracy advantage of around 4.15% over traditional techniques such as SVM, Decision Trees, and Sequential Minimal Optimisation. Python software is used to implement the suggested CNN-SVM technique. This research advances e-health systems by presenting a novel framework for effective diabetic patient monitoring that integrates machine learning, intelligent tracking, and optimisation techniques. The results point to a great deal of promise for the proposed method in the field of medicine, especially in the accurate diagnosis and follow-up of diabetic patients, which would provide opportunities for tailored and adaptable patient care.

Exploring innovative pathways for non-invasive neural communication with language interfaces, this research delves into the interdisciplinary realm of neurolinguistic learning, merging neuroscience and machine learning. It scrutinizes the intricacies of decoding neural patterns associated with language comprehension. Leveraging advanced neural network architectures, specifically Deep Recurrent Neural Networks (RNN) and Gated Recurrent Units (GRU), the study aims to amplify the landscape of neuro-device interaction. The focus of Neurolinguistic Learning lies in extracting language-related brain signals without resorting to invasive procedures. Employing cutting-edge non-invasive methods and deep learning techniques, the research aims to elevate the capabilities of neural devices such as brain-machine interfaces and neuroprosthetics. A distinctive approach involves crafting a sophisticated Deep RNN-GRU model designed to capture intricate brain patterns linked to language processing. This architectural innovation, implemented in the Python software environment, harnesses the strengths of RNNs and GRUs to enhance language decoding. The study's outcomes hold promise for advancing non-invasive brain language decoding systems, contributing to the expanding knowledge base in neurolinguistic learning. The remarkable accuracy of the proposed RNN-GRU model, boasting a 90% accuracy rate, signifies its potential application in critical real-world scenarios. This includes assistive technologies and brain-machine interfaces where precise decoding of cerebral language signals is paramount. The research underscores the efficacy of deep learning methodologies in pushing the boundaries of neurotechnology. Notably, the model outperforms established techniques, surpassing alternatives like CSP-SVM and EEGNet by an impressive 30.4% in accuracy. The model's proficiency in deciphering topic words underscores its ability to extract intricate language patterns from non-invasive brain inputs.

In programming education, code-mixed text using multiple languages or dialects simultaneously can significantly hinder learning outcomes due to misinterpretation and inadequate processing by traditional systems. For instance, students bilingual or multilingual backgrounds may face difficulties automated code reviews or multilingual coding tutorials if code-mixed queries are not accurately understood. Motivated these challenges, this paper proposes a Federated Bi-LSTM Model for Feature Extraction and Classification. This model leverages Bidirectional Long Short-Term Memory (Bi-LSTM) networks within a federated learning framework to effectively accommodate various code-switching methodologies and context dependent linguistic elements while ensuring data security privacy across distributed sources. The Federated Bi-LSTM Model demonstrates impressive performance, achieving 99.3% accuracy nearly 19% higher than traditional techniques such Support Vector Machines (SVM), Multilayer Perceptron (MLP), and Random Forest (RF). This significant improvement underscores the model's capability to efficiently analyse code mixed text and enhance programming instruction for multilingual learners. However, the model faces limitations in processing highly specialized code-mixed text and adapting to real-time applications. Future research should focus on optimizing the model for these challenges and exploring its applicability in broader domains computer-assisted education. This model represents a substantial advancement in language-aware computing, offering a promising solution for the evolving needs of adaptive and inclusive programming education technologies. This advancement has potential to transform language-sensitive computing, providing significant support for multilingual learners and setting a standard for inclusive programming education.

Federated Learning (FL), a crucial advancement in smart city technology, combines real-time traffic predictions with the potential to enhance urban mobility. This paper suggests a novel approach to real-time traffic prediction in smart cities: a hybrid Convolutional Neural Network-Recurrent Neural Network (CNN-RNN) architecture. The investigation started with the systematic collection and preprocessing of a low-resolution dataset (1.6 GB) derived from real-time Closed Circuit Television (CCTV) traffic camera images at significant intersections in Guntur and Vijayawada. The dataset has been cleaned up utilizing min-max normalization to facilitate use. The primary contribution of this study is the hybrid architecture that it develops by fusing RNN to detect temporal dynamics with CNN for geographic extraction of characteristics. While the RNN's recurrent interactions preserve hidden states for sequential processing, the CNN efficiently retrieves high-level spatial information from static traffic images. Weight adjustments and backpropagation are used in the training of the proposed hybrid model in order to enhance real-time predictions that aid in traffic management. Notably, the implementation is done with Python software. The model reaches a testing accuracy of 99.8% by the 100th epoch, demonstrating excellent performance in the results and discussion section. The Mean Absolute Error (MAE) results, which show a 4.5% improvement over existing methods like Long Short Term Memory (LSTM), Support Vector Machine (SVM), Sparse Auto Encoder (SAE), and Gated Recurrent Unit (GRU), illustrate the efficacy of the model. This demonstrates how well complex patterns may be represented by the model, yielding precise real-time traffic predictions in crowded metropolitan settings. A new era of more precise and effective real-time traffic forecasts is about to begin, thanks to the hybrid CNN-RNN architecture, which is validated by the combined strengths of FL, CNN, and RNN as well as the overall outcomes.

Predicting cyclone intensity is an important aspect of weather forecasting since it influences disaster preparation and response. This framework addresses the pressing need for precise cyclone intensity prediction by presenting a unique predictive model based on a hybrid CNN and Bi-LSTM architecture optimized using a Genetic Algorithm (GA) enhanced Fruit Fly Optimizer (FFO). Existing methods have primarily relied on traditional machine learning models and meteorological data, demonstrating limitations in capturing the complex spatial-temporal patterns inherent in cyclone evolution. These drawbacks include insufficient feature extraction abilities, underutilization of convolutional neural networks (CNN), and poor model tuning. This unique method incorporates a hybrid CNN and Bi-LSTM architecture that is tuned by a Genetic Algorithm (GA) enhanced Fruit Fly Optimizer (FFO), resulting in higher cyclone intensity prediction accuracy. The experimental results are implemented in Python software, and they reveal that this method outperforms current models by an average of 21% when compared to existing methods such as VGG-16 achieved an accuracy of 78% and Ty 5-CNN (95.23%). The suggested CNN-Bi-LSTM model predicts cyclone strength with an excellent accuracy of 99.4%. This unique approach offers a possible avenue for increasing cyclone intensity prediction, hence improving disaster preparedness and risk mitigation efforts in sensitive locations.

Remote sensing technologies, which are essential for everything from environmental monitoring to disaster relief, enable large-scale multispectral data collection. In the field of hyper-spectral imaging, where high-dimensional data is required for precise analysis, effective compression techniques are critical for transmission and storage. In the field of hyper-spectral imaging, the development of efficient compression techniques is critical because datasets containing high-dimensional information must be transmitted and stored efficiently without sacrificing analytical precision. The paper presents advanced compression techniques that combine deep Recurrent Neural Networks (RNNs) with multispectral transforms to achieve lossless compression in hyper-spectral imaging. The Discrete Wavelet Transform (DWT) is used to efficiently capture spectral and spatial information by utilizing the properties of multispectral transforms. Simultaneously, deep RNNs are used to model the hyper-spectral data with complex dependencies, allowing for sequential compression. The overall compression efficiency that is increased by the integration of spatial and spectral information allows for reduced storage requirements and improved transmission efficiency. Python software is used to implement the proposed model. When compared to Liner Spectral Mixture Analysis (LSMA) based compression, Spatial Orientation Tree Wavelet (STW)-Wavelet Difference Reduction (WDR), and DPCM, the proposed DWT-RNN-LSTM method has a better PSNR value of 45 dB and a lower MSE of 7.50%. Adaptive compression methods are presented in order to dynamically adapt to various data properties and ensure application in various hyperspectral scenes. Studies on hyper- spectral images of various sizes and resolutions demonstrate the approach's scalability and generalization, as well as the utility and adaptability of the proposed compression framework in a variety of remote sensing scenarios.

Drowsiness among drivers is a major hazard to road safety, resulting in innumerable incidents globally. Despite substantial study, existing approaches for detecting drowsiness in real time continue to confront obstacles, such as low accuracy and efficiency. In these circumstances, this study tackles the critical problems of identifying drowsiness and driver safety by suggesting a novel approach that leverages the combined effectiveness of Gated Recurrent Units (GRU) and Enhanced Deep Belief Networks (EDBN), which is optimised using COOT, a new bird collective-behavioral-based optimisation algorithm. The study begins by emphasising the relevance of sleepiness detection in improving driver safety and the limitations of prior studies in reaching high accuracy in real-time detection. The suggested method tries to close this gap by combining the GRU and EDBN simulations, which are known for their temporal modelling and feature learning capabilities, respectively, to give a comprehensive solution for sleepiness detection. Following thorough experimentation, the suggested technique achieves an outstanding accuracy of around 99%, indicating its efficiency in detecting sleepiness states in real-time driving scenarios. The relevance of this research stems from its potential to greatly reduce the number of accidents caused by drowsy driving, hence improving overall road safety. Furthermore, the use of COOT to optimize the parameters of the GRU and EDBN models adds a new dimension to the research, demonstrating the effectiveness of nature-inspired optimization methodologies for improving the performance of machine learning algorithms for critical applications such as driver safety.