Search

Car accident detection plays a crucial role in video-based traffic surveillance systems, contributing to prompt response and improved road safety. In the literature, various methods have been investigated for accident detection, among which deep learning approaches have shown superior accuracy compared to other methods. The popularity of deep learning stems from its ability to automatically learn complex features from data. However, the current research challenge in deep learning-based accident detection lies in achieving high accuracy rates while meeting real-time requirements. To address this challenge, this study introduces a deep learning approach using convolutional neural networks (CNNs) to enhance car accident detection, prioritizing accuracy and real-time performance. It includes a tailored dataset for evaluation, and the F1-scores reveal reasonably accurate detection for "damaged-rear-window" (62%) and "damaged-window" (63%), while "damaged-windscreen" exhibits exceptional performance at 83%. These results demonstrate the potential of CNNs in improving car accident detection, particularly for certain classes. Following extensive experiments and performance analysis, the proposed method demonstrates accurate results, significantly enhancing car accident detection in video-based traffic surveillance scenarios. [ABSTRACT FROM AUTHOR]

In the realm of machining, the surface finish of the final product serves as a pivotal quality indicator, signifying the excellence of the manufactured component. Consequently, a pressing requirement exists for dependable and precise predictive models that can effectively oversee the surface finish of machined parts throughout the in-process stage. This study presents a novel ensemble learning model, specifically the Convolutional Neural Network-Extreme Gradient Boosting (CNN-XG Boost), to classify the ongoing machined surface finish. To this end, a dataset containing images of machined surfaces was harnessed for training various traditional machine learning algorithms, encompassing Decision Tree (DT), Random Forest (RF), XGB, and K-Nearest Neighbors (KNN). Notably, XGB exhibited the highest accuracy at 41.6%. Expanding upon this, a deep learning CNN algorithm was trained, manifesting an elevated accuracy of 62.5% compared to its counterparts. The pinnacle of this endeavor entailed training ensemble algorithms such as CNN + DT, CNN + RF, CNN + XGB, and CNN + KNN. Among these, CNN + XGB stood out by achieving a remarkable prediction accuracy of 98%. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Abstract Cervical cancer is one of the biggest challenges in global health, thus it forms a critical need for early detection technologies that could improve patient prognosis and inform treatment decisions. This development in the form of an early detection mechanism increases the chances of successful treatment and survival, as early diagnosis promptly offers interventions that can dramatically reduce the rate of deaths attributed to this disease. Here, a customized Convolutional Neural Network (CNN) model is proposed for cervical cancerous cell detection. It includes three convolutional layers with increasing filter sizes and max-pooling layers, followed by dropout and dense layers for improved feature extraction and robust learning. By using ResNet models as inspiration, the model further innovates by incorporating skip connections into the CNN design. By enabling direct feature transmission from earlier to later layers, skip links enhance gradient flow and help preserve important spatial information. By boosting feature propagation, this integration increases the model’s ability to recognize minute patterns in cervical cell images, hence increasing classification accuracy. In our methodology, the SIPaKMeD dataset has been employed which contains 4049 cervical cell images that are arranged into five different categories. To address class imbalance, Generative Adversarial Networks (GANs) have been applied for data augmentation; that is, synthetic images have been created, that improve the diversity of the dataset and further enhance the robustness of the same. The present model is astonishingly accurate in classifying five cervical cell types: koilocytes, superficial-intermediate, parabasal, dyskeratotic, and metaplastic, thus significantly enhancing early detection and diagnosis of cervical cancer. The model gives an excellent performance because it has a validation accuracy of 99.11% and a training accuracy of 99.82%. It is a reliable model in the diagnosis of cervical cancerous cells because it ensures advancement in the computer-assisted cervical cancer detection system.

According to the World Health Organization's (WHO) external situation report on the multi-country outbreak of Monkeypox in 2023, from 11 countries in Southeast Asia Regions, Thailand recorded the highest reported cases, totaling 461. The ongoing Monkeypox outbreak has raised significant public health concerns due to its rapid spread across several nations. Early detection and diagnosis are imperative for effectively treating and controlling Monkeypox. Given this context, this study aimed to determine the most efficient model for detecting Monkeypox by employing interpretable deep learning techniques. This study utilizes deep learning techniques to diagnose Monkeypox based on images of skin lesions. We evaluate based on four models—convolutional neural network (CNN), gated recurrent unit (GRU), long short-term memory (LSTM), and bidirectional long short term memory (BiLSTM)—using a publicly available dataset. Additionally, we incorporate Local Interpretable Model-Agnostic Explanations (LIME) and techniques for explainable AI, facilitating visual interpretation of model predictions for healthcare practitioners. The CNN model's performance and LSTM model's performance have an accuracy of 100%, while the GRU model's performance and BiLSTM model's performance have an accuracy of 99.88% and 99.45%. Our findings demonstrate the effectiveness of deep learning models, including the suggested CNN model leveraging the pre-trained MobileNetV2 and LSTM. These models can play a pivotal role in combating the Monkeypox virus. Doi: 10.28991/ESJ-2024-08-05-013 Full Text: PDF

In order to solve the problems existing in the traditional frequency analysis method for the frequency analysis of grids with a high proportion of new energy, such as the large amount of calculation, the difficulty of modeling, and the prominent contradiction between the calculation speed and the calculation accuracy, this paper proposes a new frequency characteristic prediction method for the new power system based on convolutional neural network (CNN). First, the main frequency indexes of the power system with a high proportion of new energy under power disturbances are predicted using one-dimensional CNN, including the initial frequency change rate, frequency extremum, and frequency steady-state value. The prediction accuracy is improved by setting reasonable input characteristics and optimizing the parameters of the neural network. Then, the impact of disturbance location and disturbance type is further considered, and the power system characteristic data set containing disturbance information is established by the method of data dimensionality reduction. The input characteristics are constructed by using the principle of three primary channels for reference, and the extended two-dimensional CNN is used to predict the frequency security index, which improves the adaptability of CNN in the frequency analysis of grids with a high proportion of new energy. Finally, the method is verified by an example in the improved BPA 10-machine 39-node model, and the results are compared with the prediction results of the recurrent neural network, which proves that the proposed method has a high accuracy and adaptability.

This paper proposes navigation of multiple autonomous underwater vehicles (AUVs) by employing machine learning approach for wide area surveys in underwater environment. Wide area survey in underwater environment is affected by low data rate.We consider two AUVs moving in formation through clustering followed by selection of optimal path that is affected by low data rate and limited acoustical underwater communication. A state compression approach using machine learning based acoustical localization and communication (ML-ALOC) is proposed to overcome the low data rate issue in which AUV states are approximated by Hierarchical clustering followed by an optimal selection approach using Convolutional Neural Network (CNN). The performance of the proposed state compression algorithm is compared with particle state compression algorithm based on K-Means clustering at each iteration followed by Akaike information criterion (AIC) pursuing extensive simulations, in which two AUVs navigate through trajectory. It is observed from the simulations that the proposed ML-ALOC system provides better estimates when compared with acoustical localization and communication (ALOC) system using particle clustering for state compression scheme.

Neural machine translation (NMT) approaches driven by artificial intelligence (AI) has gained more and more attention in recent years, mainly due to their simplicity yet state-of-the-art performance. Despite NMT models with attention mechanism relying heavily on the accessibility of substantial parallel corpora, they have demonstrated efficacy even for languages with limited linguistic resources. The convolutional neural network (CNN) is frequently employed in tasks involving visual and speech recognition. Implementing CNN for MT is still challenging compared to the predominant approaches. Recent research has shown that the CNN-based NMT model cannot capture long-term dependencies present in the source sentence. The CNN-based model can only capture the word dependencies within the width of its filters. This unnatural character often causes a worse performance for CNN-based NMT than the RNN-based NMT models. This study introduces a simple method to improve neural translation of a low-resource language, specifically Urdu-English (UR-EN). In this paper, we use a Fully Convolutional Neural Network (FConv-NN) based NMT architecture to create a powerful MT encoder for UR-EN translation that can capture the long dependency of words in a sentence. Although the model is quite simple, it yields strong empirical results. Experimental results show that the FConv-NN model consistently outperforms the traditional CNN-based model with filters. On the Urdu-English Dataset, the FConv-NN model produces translation with a gain of 18.42 BLEU points. Moreover, the quantitative and comparative analysis shows that in a low-resource setting, FConv-NN-based NMT outperforms conventional CNN-based NMT models.

Abstract The main bottleneck in training a robust tumor segmentation algorithm for non-small cell lung cancer (NSCLC) on H&E is generating sufficient ground truth annotations. Various approaches for generating tumor labels to train a tumor segmentation model was explored. A large dataset of low-cost low-accuracy panCK-based annotations was used to pre-train the model and determine the minimum required size of the expensive but highly accurate pathologist annotations dataset. PanCK pre-training was compared to foundation models and various architectures were explored for model backbone. Proper study design and sample procurement for training a generalizable model that captured variations in NSCLC H&E was studied. H&E imaging was performed on 112 samples (three centers, two scanner types, different staining and imaging protocols). Attention U-Net architecture was trained using the large panCK-based annotations dataset (68 samples, total area 10,326 [mm2]) followed by fine-tuning using a small pathologist annotations dataset (80 samples, total area 246 [mm2]). This approach resulted in mean intersection over union (mIoU) of 82% [77 87]. Using panCK pretraining provided better performance compared to foundation models and allowed for 70% reduction in pathologist annotations with no drop in performance. Study design ensured model generalizability over variations on H&E where performance was consistent across centers, scanners, and subtypes.

Accurate prediction of sewage flow is crucial for optimizing sewage treatment processes, cutting down energy consumption, and reducing pollution incidents. Current prediction models, including traditional statistical models and machine learning models, have limited performance when handling nonlinear and high-noise data. Although deep learning models excel in time series prediction, they still face challenges such as computational complexity, overfitting, and poor performance in practical applications. Accordingly, this study proposed a combined prediction model based on an improved sparrow search algorithm (SSA), convolutional neural network (CNN), transformer, and bidirectional long short-term memory network (BiLSTM) for sewage flow prediction. Specifically, the CNN part was responsible for extracting local features from the time series, the Transformer part captured global dependencies using the attention mechanism, and the BiLSTM part performed deep temporal processing of the features. The improved SSA algorithm optimized the model's hyperparameters to improve prediction accuracy and generalization capability. The proposed model was validated on a sewage flow dataset from an actual sewage treatment plant. Experimental results showed that the introduced Transformer mechanism significantly enhanced the ability to handle long time series data, and an improved SSA algorithm effectively optimized the hyperparameter selection, improving the model's prediction accuracy and training efficiency. After introducing an improved SSA, CNN, and Transformer modules, the prediction model's $ {R^{\text{2}}} $ increased by 0.18744, $ RMSE $ (root mean square error) decreased by 114.93, and $ MAE $ (mean absolute error) decreased by 86.67. The difference between the predicted peak/trough flow and monitored peak/trough flow was within 3.6% and the predicted peak/trough flow appearance time was within 2.5 minutes away from the monitored peak/trough flow time. By employing a multi-model fusion approach, this study achieved efficient and accurate sewage flow prediction, highlighting the potential and application prospects of the model in the field of sewage treatment.

This research project aimed to address the growing concern about methane emissions from seaweed by developing a Convolutional Neural Network (CNN) model capable of accurately predicting these emissions. The study used PANDAS to read and analyze the dataset, incorporating statistical measures like mean, median, and standard deviation to understand the dataset. The CNN model was trained using the ReLU activation function and mean absolute error as the loss function. The model performance was evaluated through MAPE graphs, comparing the mean absolute percentage error (MAPE) between training and validation sets and between true and predicted emissions, and analyzing trends in yearly greenhouse gas emissions. The results demonstrated that the CNN model achieved a high level of accuracy in predicting methane emissions, with a low MAPE between the expected and actual values. This approach should enhance our understanding of methane emissions from Sargassum, contributing to more accurate environmental impact assessments and effective mitigation strategies.

Machine learning-based crop recommendation models are invaluable tools for enhancing modern AI-based farming, assisting in decisions about the selection of crops to optimize yield performance and growth. This research introduces an intelligent strategy and explainable artificial intelligence (XAI) principles based on the Convolutional Neural Network (CNN) method due to the growing demand for interpretability in modern farming decision-making, Utilizing the "Smart Agricultural Production Optimizing Engine” dataset procured from Kaggle. The proposed CNN model gives remarkable results through a comprehensive examination of soil and environmental boundaries like Nitrogen (N), Phosphorus (P), Potassium (K) levels, temperature, moistness, pH, and precipitation. Our results illustrate that the proposed framework essentially moves forward the precision of trim suggestions, advertising a promising arrangement for modernizing agricultural practices and guaranteeing maintainable crop yields.

Anomaly detection is one of the many challenging areas in cybersecurity. The anomaly can occur in many forms, such as fraudulent credit card transactions, network intrusions, and anomalous imageries or documents. One of the most common challenges in anomaly detection is the obscurity of the normal state and the lack of anomalous samples. Traditionally, this problem is tackled by using resampling techniques or choosing models that approximate the distribution of the normal states. Variational AutoEncoder (VAE) has been studied in anomaly detections despite being more suitable in generative tasks. This study aims to explore the usage of VAE in credit card anomaly detection and evaluate latent space sampling techniques. In this study, we evaluate the usage of the convolutional network-based VAE model on a credit card transaction dataset. We train two VAE models, one with a large number of normal data and one with a small number of anomalous data. We compare the performance of both VAE models and evaluate the latent space of both VAE models by rescaling them with reconstruction error vectors. We also compare the effectiveness of the VAE model with other anomaly detection models when they are trained on imbalanced dataset.

Abstract To ensure the reliability of machining quality, it is crucial to predict tool wear accurately. In this paper, a novel deep learning-based model is proposed, which synthesizes the advantages of power spectral density (PSD), convolutional neural networks (CNN), and vision transformer model (ViT), namely PSD-CVT. PSD maps can provide a comprehensive understanding of the spectral characteristics of the signals. It makes the spectral characteristics more obvious and makes it easy to analyze and compare different signals. CNN focuses on local feature extraction, which can capture local information such as the texture, edge, and shape of the image, while the attention mechanism in ViT can effectively capture the global structure and long-range dependencies present in the image. Two fully connected layers with a ReLU function are used to obtain the predicted tool wear values. The experimental results on the PHM 2010 dataset demonstrate that the proposed model has higher prediction accuracy than the CNN model or ViT model alone, as well as outperforms several existing methods in accurately predicting tool wear. The proposed prediction method can also be applied to predict tool wear in other machining fields.

The objective of super-resolution in remote sensing imagery is to enhance low-resolution images to recover high-quality details. With the rapid progress of deep learning technology, the deep learning-based super-resolution technology for remote sensing images has also made remarkable achievements. However, these methods encounter several challenges. They often struggle with processing long-range spatial information that encompasses complex scene changes, adversely affecting the image's coherence and accuracy. Furthermore, the lack of connectivity in feature extraction blocks hinders effective feature utilization in deeper network layers, leading to issues such as gradient vanishing and exploding. Additionally, constraints in the spatial domain of previous methods frequently result in severe shape distortion and blurring. To address these issues, this study proposes the CFFormer, a new super-resolution framework that employs the Swin Transformer as its core architecture and incorporates the Channel Fourier Block (CFB) to refine features in the frequency domain. The Global Attention Block (GAB) is also integrated to enhance global information capture, thereby improving the extraction of spatial features. To increase model stability and feature utilization efficiency, a Jump-Joint Fusion Mechanism is designed, culminating in a Residual Fusion Swin Transformer Block (RFSTB) that alleviates the gradient vanishing issue and optimizes feature reuse. Experimental results confirm the CFFormer's superior performance in remote sensing image reconstruction, demonstrating outstanding perceptual quality and reliability. Notably, the CFFormer achieves a Peak Signal-to-Noise Ratio (PSNR) of 29.83 dB on the UcMercedx4 dataset, surpassing the SwinIR method by approximately 0.5 dB, indicating a substantial enhancement.

In the evolving digital landscape, interconnected IoT networks are expanding fast. However, essential security measures are often lacking, which makes it vulnerable to cyber threats. Intrusion Detection Systems (IDSs) are crucial in combating these threats, but IDSs in the IoT domain face significant challenges; one of them is the existence of imbalanced data, where attack activities are underrepresented compared to normal activities. This becomes a problem when handling Machine Learning (ML) tasks to improve the system’s intrusion detection ability. This paper presents a novel solution by leveraging Generative Adversarial Networks (GANs) to create balanced datasets. Unlike traditional methods, our approach generates high-quality synthetic samples that maintain the intricate characteristics of the data, making it possible to improve detection accuracy. Additionally, we introduce three specialized IDSs designed to detect attacks on the lightweight Message Queuing Telemetry Transport (MQTT) protocol, utilizing hybrid Deep Learning (DL) algorithms (CNN-RNN, CNN-LSTM, and CNN-GRU). Experimental results, using a combination of both the original MQTT dataset and a newly generated dataset through GANs, demonstrate the better performance of the generated datasets in multi-class classification. This performance highlights the effectiveness of our GAN-based approach in improving detection accuracy at the same time that it reduces false positives. The ultimate objective is to develop an autonomous system tailored to safeguard IoT environments, providing a robust and adaptable solution to emerging cyber threats. [ABSTRACT FROM AUTHOR]

In this paper, a frequency selective surface (FSS) for n41, n78, n257, n258, n260, and n261 bands electromagnetic interference (EMI) shielding is proposed using a deep learning method. The size of the FSS unit cell is 4.8 mm (0.071 λ 0 ), where λ 0 is the wavelength corresponding to the cutoff frequency of the −10 dB stopband where the n41 and n78 bands are located. This FSS exhibits three key characteristics. Firstly, it possesses ultrawideband capability, with a stopband frequency range of 23.922–40.023 GHz and a relative bandwidth reaching 50.4%. Secondly, it effectively shields against electromagnetic radiation in multiple 5 G bands, including n41, n78, n257, n258, n260, and n261. Thirdly, it demonstrates high selectivity, with a very narrow transition band of only 0.405 GHz between the passband and the stopband. To validate the designed FSS, a prototype of this structure was first fabricated for experimentation. The experimental results show that the measured S21 is highly consistent with the simulated S21. Secondly, the unit cell array of FSS is placed above the monopole antenna for co-simulation. It can be observed that after placing the FSS, the intensity of electromagnetic radiation propagating in space is significantly reduced by up to more than 90%. Therefore, the proposed FSS can be effectively applied for EMI shielding in 5 G bands. [ABSTRACT FROM AUTHOR]

The paper proposes novel approaches to achieve rapid and accurate inference of aerodynamic performance parameters for airfoil optimization design using deep learning methods, compared with expensive computational fluid dynamics (CFD) simulation tools. Two deep neural networks are proposed, and they were tested for aerodynamic performance prediction (APPN), including the lift and drag coefficients and the surface pressure distribution. The network framework is constructed using a convolutional neural network (CNN) and a fully connected neural network (FCN). To enhance the prediction capability of the two networks under varying operating conditions including various angles of attack (AOAs), Mach numbers, and Reynolds numbers (R), a flexible and practical parameterization method called the signed distance function (SDF) is utilized to represent the geometric information and then merged with the Reynolds number for input to the network models. The prediction results show that the proposed models have clear advantages, including fast convergence, high prediction accuracy, and remarkable extensionality. Moreover, the predicted aerodynamic coefficients of an airfoil can be obtained within 13 ms, 4 orders of magnitudes faster than CFD solvers. The prediction relative errors of the lift and drag coefficients are lower than 1.5%, and that of the pressure distribution is less than 1.2%. This work indicates that deep learning–based network models can offer great potential as practical tools for aerodynamic design and optimization and for flow control in the future. [ABSTRACT FROM AUTHOR]

In the era of rapidly advancing medical technologies, the segmentation of medical data has become inevitable, necessitating the development of privacy preserving machine learning algorithms that can train on distributed data. Consolidating sensitive medical data is not always an option particularly due to the stringent privacy regulations imposed by the Health Insurance Portability and Accountability Act. In this paper, I introduce a HIPAA compliant framework that can train from distributed data. I then propose a multimodal vertical federated model for Alzheimer's disease detection, a serious neurodegenerative condition that can cause dementia, severely impairing brain function and hindering simple tasks, especially without preventative care. This vertical federated learning model offers a distributed architecture that enables collaborative learning across diverse sources of medical data while respecting privacy constraints imposed by HIPAA. The VFL architecture proposed herein offers a novel distributed architecture, enabling collaborative learning across diverse sources of medical data while respecting statutory privacy constraints. By leveraging multiple modalities of data, the robustness and accuracy of AD detection can be enhanced. This model not only contributes to the advancement of federated learning techniques but also holds promise for overcoming the hurdles posed by data segmentation in medical research. [ABSTRACT FROM AUTHOR]

The most common type of malignant brain tumor, gliomas, has a variety of grades that significantly impact a patient's chance of survival. Accurate segmentation of brain tumor regions from MRI images is crucial for enhancing diagnostic precision and refining surgical strategies. This task is particularly challenging due to the diverse sizes and shapes of tumors, as well as the intricate nature of MRI data. Mastering this segmentation process is essential for improving clinical outcomes and ensuring optimal treatment planning. In this research, we provide a UNet-based model (RR-UNet) designed specifically for brain tumor segmentation, which uses small and diverse datasets containing human-annotated ground truth segmentations. This model uses residual learning to improve segmentation results over the original UNet architecture, as shown by higher dice similarity coefficient (DSC) and Intersection over Union (IoU) scores. Residual blocks enable a deeper network, which can capture complex patterns. Residual blocks reuse features, allowing the network to learn more abstract and informative representations from input images. Through comprehensive evaluation and validation, we illustrate our method's efficacy and generalization capabilities, emphasizing its potential for real-world clinical applications. This segmentation model predicts DSC of 98.18% and accuracy of 99.78% in tumor segmentation using Figshare LGG (Low-grade glioma) FLAIR segmentation dataset and DSC of 98.54% & accuracy of 99.81% using BraTS 2020 dataset. The ablation study shows the importance of the model's residual mechanism. Overall, the proposed approach outperforms or compares to existing most recent algorithms in brain tumor segmentation tasks. [ABSTRACT FROM AUTHOR]

Phishing attacks are the biggest cybersecurity threats in the digital world. Attackers exploit users by impersonating real, authentic websites to obtain sensitive information such as passwords and bank statements. One common technique in these attacks is using malicious URLs. These malicious URLs mimic legitimate URLs, misleading users into interacting with malicious websites. This practice, URL phishing, presents a big threat to internet security, emphasizing the need for advanced detection methods. So we aim to enhance phishing URL detection by using machine learning and deep learning models, leveraging a set of low-level URL features derived from n-gram analysis. In this paper, we present a method for detecting malicious URLs using statistical features extracted from n-grams. These n-grams are extracted from the hexadecimal representation of URLs. We employed 4 experiments in our paper. The first 3 experiments used machine learning with the statistical features extracted from these n-grams, and the fourth experiment used these grams directly with deep learning models to evaluate their effectiveness. Also, we used Explainable AI (XAI) to explore the extracted features and evaluate their importance and role in phishing detection. A key advantage of our method is its ability to reduce the number of features required and reduce the training time by using fewer features after applying XAI techniques. This stands in contrast to the previous study, which relies on high-level URL features and needs pre-processing and a high number of features (87 high-level URL-based features). So our technique only uses statistical features extracted from n-grams and the n-gram itself, without the need for any high-level features. Our method is evaluated across different n-gram lengths (2, 4, 6, and 8), aiming to optimize detection accuracy. We conducted four experiments in our study. In the first experiment, we focused on extracting and using 12 common statistical features like mean, median, etc. In the first experiment, the XGBoost model achieved the highest accuracy using 8-gram features with 82.41%. In the second experiment, we expanded the feature set and extracted an additional 13 features, so our feature count became 25. XGBoost in the second experiment achieved the highest accuracy with 86.40%. Accuracy improvement continued in the third experiment, we extracted an additional 16 features (character count features), and these features increased XGBoost accuracy to 88.15% in the third experiment. In the fourth experiment, we directly fed n-gram representations into deep learning models. The Convolutional Neural Network (CNN) model achieved the highest accuracy of 94.09% in experiment four. Also, we applied XAI techniques, SHapley Additive exPlanations (SHAP), and Local Interpretable Model-agnostic Explanations (LIME). Through the explanation provided by XAI methods, we were able to determine the most important features in our feature set, enabling a reduction in feature count. Using fewer features (4, 7, 10, 13, 15), we got good accuracy compared to the 41 features used in experiment three and reduced the models' training times and complexity. This research aimed to enhance phishing URL detection by using machine learning and deep learning models, leveraging a set of low-level URL features derived from n-gram analysis. Our findings show the importance of using minimal statistical features to identify malicious URLs. Notably, the use of CNN had a great advancement, achieving an accuracy rate of 94.09% with using n-grams of URLs, surpassing traditional machine learning models. This achievement not only validates the efficacy of deep learning models in complex pattern recognition tasks but also highlights the efficiency of our feature selection approach, which relies on a lower number of features and is less complex compared to existing high-level feature-based studies. The research outcomes demonstrate a promising pathway toward developing more robust, efficient, and scalable phishing detection systems. [ABSTRACT FROM AUTHOR]

A cardiac arrhythmia is an abnormal heart rhythm caused by irregular heartbeats. Cardiac arrhythmias include atrial or ventricular fibrillation, right or left bundle branch block beats, and premature atrial or ventricular contractions. Different cardiac arrhythmias have distinct causes and clinical presentations. The type of cardiac arrhythmia must be identified to enable further intervention and treatment for addressing its underlying causes. In this study, we developed a convolutional neural network (CNN) model that extracts and classifies time-domain features to detect cardiac arrhythmias automatically in electrocardiogram (ECG) signals. This model employs endpoint detection to detect the activity of time-domain signals in accordance with a threshold for identifying the peak wave in ECG signals. These features are then transferred to two-dimensional (2D) color patterns that indicate abnormal heartbeats. Subsequently, a one-dimensional (1D) or 2D CNN classifier is employed to distinguish normal heartbeats from cardiac arrhythmias in raw ECG data. The proposed model was trained, tested, and validated on the Massachusetts Institute of Technology–Beth Israel Deaconess Medical Center Arrhythmia Database (commonly known as the MIT-BIH Arrhythmia Database), and it exhibited promising performance in cardiac arrhythmia recognition, as indicated by its precision, recall, F1 score, and accuracy. [ABSTRACT FROM AUTHOR]

Copyright of Baghdad Science Journal is the property of Republic of Iraq Ministry of Higher Education & Scientific Research (MOHESR) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In this study, a nanofluid ultrasonic atomization minimum quantity lubrication (MQL) system was employed for grinding and processing titanium alloy (Ti-6Al-4V). Nanoparticles in the nanofluid were dispersed through the ultrasonic atomization MQL system to improve the lubrication efficiency. This system is a grinding manufacturing innovation on lubrication technology. First, a robust design method optimizes the grinding parameters of force ratio, temperature, and surface roughness of each quality characteristic. Surface integrity is considered a key component metric in grinding. Therefore, various images of the grinding surface quality were collected to establish a complete titanium alloy grinding image database. The convolutional neural network (CNN) technique was used to establish a model with high accuracy and automatic recognition of the surface quality after grinding, which is classified as Good, Plowing, Crushing pits, and Bad. This study employed VGG-19, GoogLeNet, ResNet-50, and AlexNet for modeling and testing comparison. Different hyperparameter configurations will affect each model's learning speed and performance. Therefore, their hyperparameter optimizations were performed before establishing our CNN model for classification. After optimization, the average accuracy of the four models can be improved by 3.75–12%, which can quickly improve the performance of the CNN model systematically. This study verified and validated our model. The findings show that this research method can be used by the industry in the actual grinding process monitoring and identification of the grinding quality of Ti-6Al-4V. [ABSTRACT FROM AUTHOR]

Deep learning-powered AI tools offer significant potential to improve COVID-19 lung infection diagnosis. This study proposes a novel AI-based image analysis method for multiclass classification. We analyzed publicly available datasets from Italian Society of Medical and Interventional Radiology (SIRM), Kaggle, and Radiopaedia. However, the relevance, strength, and relationships of static features extracted from these images require further investigation. Bayesian inference approaches have recently emerged as powerful tools for analyzing static features. These approaches can reveal hidden dynamics and relationships between features. Using Analysis of variance (ANOVA) based ranking techniques, we extracted gray level co-occurrence matrix (GLCM) features from images belonging to three classes such as COVID-19, bacterial pneumonia, and normal. To delve deeper into the dynamic behavior and optimize its diagnostic potential, Homogeneity (identified as the most significant feature) was chosen for further analysis using dynamic profiling and optimization methods. This focused investigation aimed to decipher the intricate, non-linear dynamics within GLCM features across all three classes. Our method offers a two-fold benefit. First, it deepens our understanding of the intricate relationships between features extracted from chest X-rays using gray level co-occurrence matrix analysis. Second, it provides a comprehensive examination of these features themselves. This combined analysis sheds light on the hidden dynamics that are crucial for accurate diagnosis and prognosis of various infectious diseases. In addition to the above, we have developed a novel AI-powered imaging analysis method for multiclass classification. This innovative approach has the potential to significantly improve diagnostic accuracy and prognosis of infectious diseases, particularly COVID-19. [ABSTRACT FROM AUTHOR]

Accurate traffic volume prediction is crucial for intelligent transportation systems to control traffic conditions and improve travel efficiency. Traditional traffic volume prediction models focus on the similarity of passenger volume patterns in historical data. However, they ignore the continuous and periodic features of traffic volume data and the deviation in traffic volume caused by external factors such as holidays and weather. This paper proposes a multifeature fusion convolutional long-short-term memory (CLSTM) model. The model is based on a convolutional neural network (CNN) and a long-short-term memory (LSTM) neural network. The CLSTM model considers time continuity as a short-term feature, daily periodicity as a long-term feature, spatial correlation between roads as a spatial feature, and environmental factors as external features. The CNN model is applied to represent the temporal and spatial features as a two-dimensional spatial-temporal matrix, and two sets of high-level features are proposed. The fully connected neural network model is used to fuse the predictions from the feature matrix and LSTM neural networks. The effectiveness of feature extraction, model design, and model sensitivity are tested using the London M25 motorway as the research object. The results illustrate that the CLSTM model enhances both prediction accuracy and model adaptability, achieving a balance between prediction efficiency and accuracy. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Shanghai Jiaotong University (Science) is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Bridges are crucial components of infrastructure networks that facilitate national connectivity and development. According to the National Bridge Inventory (NBI) and the Federal Highway Administration (FHWA), the cost to repair U.S. bridges was recently estimated at approximately USD 164 billion. Traditionally, bridge inspections are performed manually, which poses several challenges in terms of safety, efficiency, and accessibility. To address these issues, this research study introduces a method using Unmanned Aerial Systems (UASs) to help automate the inspection process. This methodology employs UASs to capture visual images of a concrete bridge deck, which are then analyzed using advanced machine learning techniques of Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) to detect damage and delamination. A case study on the Beyer Road Concrete Bridge in Michigan is used to demonstrate the developed methodology. The findings demonstrate that the ViT model outperforms the CNN in detecting bridge deck damage, with an accuracy of 97%, compared to 92% for the CNN. Additionally, the ViT model showed a precision of 96% and a recall of 97%, while the CNN model achieved a precision of 93% and a recall of 61%. This technology not only enhances the maintenance of bridges but also significantly reduces the risks associated with traditional inspection methods.

Abstract Sandification can degrade the strength and quality of dolomite, and to a certain extent, compromise the stability of a tunnel's surrounding rock as an unfavorable geological boundary. Sandification degree classification of sandy dolomite is one of the non-trivial challenges faced by geotechnical engineering projects such as tunneling in complex geographical environments. The traditional methods quantitatively measuring the physical parameters or analyzing some visual features are either time-consuming or inaccurate in practical use. To address these issues, we, for the first time, introduce the convolutional neural network (CNN)-based image classification methods into dolomite sandification degree classification task. In this study, we have made a significant contribution by establishing a large-scale dataset comprising 5729 images, classified into four distinct sandification degrees of sandy dolomite. These images were collected from the vicinity of a tunnel located in the Yuxi section of the CYWD Project in China. We conducted comprehensive classification experiments using this dataset. The results of these experiments demonstrate the groundbreaking achievement of CNN-based models, which achieved an impressive accuracy rate of up to 91.4%. This accomplishment underscores the pioneering role of our work in creating this dataset and its potential for applications in complex geographical analyses.

Abstract Epilepsy is one of the most well-known neurological disorders globally, leading to individuals experiencing sudden seizures and significantly impacting their quality of life. Hence, there is an urgent necessity for an efficient method to detect and predict seizures in order to mitigate the risks faced by epilepsy patients. In this paper, a new method for seizure detection and prediction is proposed, which is based on multi-class feature fusion and the convolutional neural network-gated recurrent unit-attention mechanism (CNN-GRU-AM) model. Initially, the Electroencephalography (EEG) signal undergoes wavelet decomposition through the Discrete Wavelet Transform (DWT), resulting in six subbands. Subsequently, time–frequency domain and nonlinear features are extracted from each subband. Finally, the CNN-GRU-AM further extracts features and performs classification. The CHB-MIT dataset is used to validate the proposed approach. The results of tenfold cross validation show that our method achieved a sensitivity of 99.24% and 95.47%, specificity of 99.51% and 94.93%, accuracy of 99.35% and 95.16%, and an AUC of 99.34% and 95.15% in seizure detection and prediction tasks, respectively. The results show that the method proposed in this paper can effectively achieve high-precision detection and prediction of seizures, so as to remind patients and doctors to take timely protective measures.

With the continuous advancement of Internet of Things (IoT) intelligence, IoT security issues have become more and more prominent in recent years. The research on IoT security has become a hot spot. A lightweight IoT intrusion detection model fusing a convolutional neural network, bidirectional long short-term memory network is proposed. It aims to improve processed data security and attack detection accuracy. First, sampling is performed by a hybrid sampling algorithm fusing SMOTE and ENN. Its aim is to minimize the impact of imbalanced-data and ensure data quantity in the process. Then, the data features are extracted by 2-dimensional convolutional neural network (2dCNN), and the effect of useless information is reduced by mean pooling and maximum pooling, so it can be adapted to the demanding resource environment of the IoT. On this basis, long-range dependent temporal features are extracted using bidirectional long short-term memory (BiLSTM), which aims to fully extract data features to improve detection accuracy in the limited resource environment. Finally, the algorithm is validated on the UNSW_NB15 dataset, and the results of the experiments reaches 93.5% at Accuracy, 86.4% at Precision, 85.3% at Recall and 85.8% at F1-Score. According to the results, the proposed algorithm can generate higher-quality samples, achieve higher detection rate with faster inference time and spend lower memory costs. This paper is part of special issue AI-DRIVEN SECURE COMMUNICATION IN MASSIVE IOT FOR 5G AND BEYOND.

Recent advances in artificial intelligence have opened up new avenues for microstructure characterization, notably in metallic materials. Physical and mechanical properties generally depend on the microstructure of the metallic material. On the other hand, microstructural characterization takes time and calls for specific techniques that don’t always lead to conclusive results quickly. To address this issue, this research focuses on the application of artificial intelligence approaches to microstructural categorization. We demonstrate the advantages of the AI approach using an example of Al-Si alloy, a material that is widely employed in a variety of industries. To specify a suitable convolutional neural network (CNN) approach for the microstructural classification of the Al-Si alloy, CNN models were trained and compared using DenseNet201, Inception v3, InceptionResNetV2, ResNet152V2, VGG16, and Xception architectures. Resulting from the comparison, it was determined that the developed supervised transfer learning model can execute the microstructural classification of Al-Si alloy microstructural images. This paper is an attempt to advance methods of microstructure recognition/classification/characterization by using Deep Learning approaches. The significance of the established model is demonstrated and its accordance with the literature data. Also, necessity is shown of developing material models and optimization through systematic microstructural investigation, production conditions, and material attributes.

Abstract The use of deep learning for identifying defects in medical images has rapidly emerged as a significant area of interest across various medical diagnostic applications. The automated recognition of Posterior Fossa Tumors (PFT) in Magnetic Resonance Imaging (MRI) plays a vital role, as it furnishes essential data about irregular tissue, essential for treatment planning. Human examination has traditionally been the standard approach for identifying defects in brain MRI. This technique is unsuitable for a massive quantity of data. Therefore, automated PFT detection techniques are being established to minimize radiologist's time. In this paper, the posterior fossa tumor is detected and classified in brain MRI using Convolutional Neural Network (CNN) algorithms, and the model result and accuracy obtained from each algorithm are explained. A dataset collection made up of 3,00,000 images with an average of 500 patients from the Children's Cancer Hospital Egypt (CCHE) was used. The CNN algorithms investigated to classify the PFT were VGG19, VGG16, and ResNet50. Moreover, explanations for the behavior of networks were investigated using three different techniques: LIME, SHAP, and ICE. Overall, the results showed that the best model was VGG16 compared with other CNN-used models with accuracy rate values of 95.33%, 93.25%, and 87.4%, respectively.

Geomechanical assessment using coupled reservoir-geomechanical simulation is becoming increasingly important for analyzing the potential geomechanical risks in subsurface geological developments. However, a robust and efficient geomechanical upscaling technique for heterogeneous geological reservoirs is lacking to advance the applications of three-dimensional (3D) reservoir-scale geomechanical simulation considering detailed geological heterogeneities. Here, we develop convolutional neural network (CNN) proxies that reproduce the anisotropic nonlinear geomechanical response caused by lithological heterogeneity, and compute upscaled geomechanical properties from CNN proxies. The CNN proxies are trained using a large dataset of randomly generated spatially correlated sand-shale realizations as inputs and simulation results of their macroscopic geomechanical response as outputs. The trained CNN models can provide the upscaled shear strength (R2 > 0.949), stress-strain behavior (R2 > 0.925), and volumetric strain changes (R2 > 0.958) that highly agree with the numerical simulation results while saving over two orders of magnitude of computational time. This is a major advantage in computing the upscaled geomechanical properties directly from geological realizations without the need to perform local numerical simulations to obtain the geomechanical response. The proposed CNN proxy-based upscaling technique has the ability to (1) bridge the gap between the fine-scale geocellular models considering geological uncertainties and computationally efficient geomechanical models used to assess the geomechanical risks of large-scale subsurface development, and (2) improve the efficiency of numerical upscaling techniques that rely on local numerical simulations, leading to significantly increased computational time for uncertainty quantification using numerous geological realizations.

We have studied a strongly nonlinear backward stochastic partial differential equation (B-SPDE) through an approximation method and with machine learning (ML)-based Monte Carlo simulation. This equation is well-known and was previously derived from studies in finance. However, how to analyze and solve this equation has remained a problem for quite a long time. The main difficulty is due to the singularity of the B-SPDE since it is a strongly nonlinear one. Therefore, by introducing new truncation operators and integrating the machine learning technique into the platform of a convolutional neural network (CNN), we have developed an effective approximation method with a Monte Carlo simulation algorithm to tackle the well-known open problem. In doing so, the existence and uniqueness of a 2-tuple adapted strong solution to an approximation B-SPDE were proved. Meanwhile, the convergence of a newly designed simulation algorithm was established. Simulation examples and an application in finance were also provided.

Considering the problem of radar target detection in the sea clutter environment, this paper proposes a deep learning-based marine target detector. The proposed detector increases the differences between the target and clutter by fusing multiple complementary features extracted from different data sources, thereby improving the detection performance for marine targets. Specifically, the detector uses two feature extraction branches to extract multiple levels of fast-time and range features from the range profiles and the range-Doppler (RD) spectrum, respectively. Subsequently, the local-global feature extraction structure is developed to extract the sequence relations from the slow time or Doppler dimension of the features. Furthermore, the feature fusion block is proposed based on adaptive convolution weight learning to efficiently fuse slow-fast time and RD features. Finally, the detection results are obtained through upsampling and nonlinear mapping to the fused multiple levels of features. Experiments on two public radar databases validated the detection performance of the proposed detector.

Summary: Among the recent network security issues, Distributed Denial of Service (DDoS) attack is one of the most dangerous threats in today's cyberspace that can disrupt essential services. These attacks compromise network security by flooding the target with malicious traffic. Several researchers have designed effective DDoS detection mechanisms using machine learning (ML) and deep learning (DL)‐based techniques. However, existing detection approaches paid less attention to issues such as class imbalance, multi‐classification, or computational cost of the models, especially time. In this study, we propose a novel framework for detecting and classifying DDoS attacks with high accuracy and low computational cost. To address the class imbalance, we employ random sampling, while feature selection techniques such as low information gain, quasi‐constant elimination, and principal component analysis are utilized for optimal feature selection and reduction. Our proposed CNN‐based model achieves outstanding performance, boasting an accuracy of 99.99% for binary classification and 98.44% for multi‐classification. The proposed model is compared to existing works and baseline line models and found to be effective in binary and multi‐classification. [ABSTRACT FROM AUTHOR]

Skin cancer is a dreadful disease, which is mainly caused due to the heavy exposure of the human body to the ultraviolet rays emitted from the sun. Although the mortality rate is very high, the survival rate is found to be superior when it is detected at its early stage. In this research, Chronological Dingo Optimizer (CDO)-based Deep Maxout Network (DMN) is developed for the skin cancer detection. Here, initially the images are pre-processed and then, the skin lesion segmentation is accomplished effectively by employing Double U-Net. Then, data augmentation is performed and the detection process is carried out by employing DMN, where the network is optimally fine-tuned utilizing designed CDO. The CDO is the integration of chronological concept with Dingo Optimizer (DOX). The proposed scheme shows better outcomes with superior sensitivity of 0.959, F-measure of 0.908, accuracy of 0.923 and specificity of 0.837. [ABSTRACT FROM AUTHOR]

Neurodegenerative disorders, notably Alzheimer's, pose an escalating global health challenge. Marked by the degeneration of brain neurons, these conditions lead to a gradual decline in nerve cells. Worldwide, over 55 million people grapple with dementia, with Alzheimer's prominently impacting the aging demographic. The primary hurdle to early Alzheimer's detection is the widespread lack of awareness. The main goal is to design and implement an artificial intelligence system using deep learning (DL) to detect Alzheimer's disease (AD) through medical images and classify them into various stages, such as non-demented, moderate dementia, mild dementia, and very mild dementia. The dataset contains 6400 magnetic resonance images in .jpg format, with standardized dimensions of 176 × 208 pixels. To demonstrate the advantages of data augmentation and transformation techniques, four scenarios were created: two without these techniques, utilizing the Adam and SGD optimizers, and two with these techniques, also employing the Adam and SGD optimizers, respectively. The main results revealed that scenarios utilizing these techniques exhibited more stable performance when validated with a new dataset. Scenario 3, using the Adam optimizer, achieved a weighted average accuracy of 91.83%, whereas scenario 4, employing the SGD optimizer, reached 87.58% accuracy. In contrast, scenarios 1 and 2, which omitted these techniques, obtained low accuracies below 55%. It is concluded that classifying AD with a DL model exceeding 90% accuracy is feasible. This is the importance of utilizing data augmentation and transformation techniques to improve generalizability to input image variations, which is a consistent factor in the healthcare sector. [ABSTRACT FROM AUTHOR]

Sign language is undoubtedly a common way of communication among deaf and non-verbal people. But it is not common among hearing people to use sign language to express feelings or share information in everyday life. Therefore, a significant communication gap exists between deaf and hearing individuals, despite both groups experiencing similar emotions and sentiments. In this paper, we developed a convolutional neural network–squeeze excitation network to predict the sign language signs and developed a smartphone application to provide access to the ML model to use it. The SE block provides attention to the channel of the image, thus improving the performance of the model. On the other hand, the smartphone application brings the ML model close to people so that everyone can benefit from it. In addition, we used the Shapley additive explanation to interpret the black box nature of the ML model and understand the models working from within. Using our ML model, we achieved an accuracy of 99.86% on the KU-BdSL dataset. The SHAP analysis shows that the model primarily relies on hand-related visual cues to predict sign language signs, aligning with human communication patterns. [ABSTRACT FROM AUTHOR]

Radar signal intra-pulse modulation recognition can be addressed with convolutional neural networks (CNNs) and time–frequency images (TFIs). However, current CNNs have high computational complexity and do not perform well in low-signal-to-noise ratio (SNR) scenarios. In this paper, we propose a lightweight CNN known as the cross-scale aware network (CSANet) to recognize intra-pulse modulation based on three types of TFIs. The cross-scale aware (CSA) module, designed as a residual and parallel architecture, comprises a depthwise dilated convolution group (DDConv Group), a cross-channel interaction (CCI) mechanism, and spatial information focus (SIF). DDConv Group produces multiple-scale features with a dynamic receptive field, CCI fuses the features and mitigates noise in multiple channels, and SIF is aware of the cross-scale details of TFI structures. Furthermore, we develop a novel time–frequency fusion (TFF) feature based on three types of TFIs by employing image preprocessing techniques, i.e., adaptive binarization, morphological processing, and feature fusion. Experiments demonstrate that CSANet achieves higher accuracy with our TFF compared to other TFIs. Meanwhile, CSANet outperforms cutting-edge networks across twelve radar signal datasets, providing an efficient solution for high-precision recognition in low-SNR scenarios. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Sandification can degrade the strength and quality of dolomite, and to a certain extent, compromise the stability of a tunnel's surrounding rock as an unfavorable geological boundary. Sandification degree classification of sandy dolomite is one of the non-trivial challenges faced by geotechnical engineering projects such as tunneling in complex geographical environments. The traditional methods quantitatively measuring the physical parameters or analyzing some visual features are either time-consuming or inaccurate in practical use. To address these issues, we, for the first time, introduce the convolutional neural network (CNN)-based image classification methods into dolomite sandification degree classification task. In this study, we have made a significant contribution by establishing a large-scale dataset comprising 5729 images, classified into four distinct sandification degrees of sandy dolomite. These images were collected from the vicinity of a tunnel located in the Yuxi section of the CYWD Project in China. We conducted comprehensive classification experiments using this dataset. The results of these experiments demonstrate the groundbreaking achievement of CNN-based models, which achieved an impressive accuracy rate of up to 91.4%. This accomplishment underscores the pioneering role of our work in creating this dataset and its potential for applications in complex geographical analyses. [ABSTRACT FROM AUTHOR]

The Global Ecosystem Dynamics Investigation (GEDI) LiDAR data have been widely used to measure tree canopy heights. However, due to the altered GEDI waveform geometry in the fire-consumed canopy layers, it is challenging to accurately detect diminished signals with the original GEDI threshold settings, which may lead to underestimation of canopy height under extreme fire conditions. In this study, we delve into the relationship between the accuracy of post-fire GEDI canopy height and front thresholds applied to GEDI waveforms. Notably, thresholds 3σ and 6σ from the GEDI settings proved too high to detect burnt canopy signals, resulting in the lowest accuracy and substantial underestimation. In contrast, the 1σ threshold achieved the highest accuracy. Subsequently, a 1D CNN model was developed to calibrate post-fire GEDI canopy height, utilizing L1B waveforms as input, resulting in a significant reduction of RMSE from 12.2 m to 6.7 m and the elimination of canopy height underestimation in high and extreme fire severity scenarios. In the third experiment, a 2D CNN model was constructed for mapping calibrated canopy height with Sentinel-2 images, revealing a noteworthy drop in RMSE from 14.6 m to 8.5 m compared to the same model using original GEDI canopy height. [ABSTRACT FROM AUTHOR]

Maintaining security in communication networks has long been a major concern. This issue has become increasingly crucial due to the emergence of new communication architectures like the Internet of Things (IoT) and the advancement and complexity of infiltration techniques. For usage in networks based on the Internet of Things, previous intrusion detection systems (IDSs), which often use a centralized design to identify threats, are now ineffective. For the resolution of these issues, this study presents a novel and cooperative approach to IoT intrusion detection that may be useful in resolving certain current security issues. The suggested approach chooses the most important attributes that best describe the communication between objects by using Black Hole Optimization (BHO). Additionally, a novel method for describing the network's matrix-based communication properties is put forward. The inputs of the suggested intrusion detection model consist of these two feature sets. The suggested technique splits the network into a number of subnets using the software-defined network (SDN). Monitoring of each subnet is done by a controller node, which uses a parallel combination of convolutional neural networks (PCNN) to determine the presence of security threats in the traffic passing through its subnet. The proposed method also uses the majority voting approach for the cooperation of controller nodes in order to more accurately detect attacks. The findings demonstrate that, in comparison to the prior approaches, the suggested cooperative strategy can detect assaults in the NSLKDD and NSW-NB15 datasets with an accuracy of 99.89 and 97.72 percent, respectively. This is a minimum 0.6 percent improvement. [ABSTRACT FROM AUTHOR]

Spectral confusion among land cover classes is quite common, let alone in a complex and heterogenous system like the semi-arid Mediterranean environment; thus, employing new developments in remote sensing, such as multispectral imagery (MSI) captured by unmanned aerial vehicles (UAVs) and airborne light detection and ranging (LiDAR) techniques, with deep learning (DL) algorithms for land cover classification can help to address this problem. Therefore, we propose an image-based land cover classification methodology based on fusing multispectral and airborne LiDAR data by adopting CNN-based semantic segmentation in a semi-arid Mediterranean area of northeastern Aegean, Greece. The methodology consists of three stages: (i) data pre-processing, (ii) semantic segmentation, and (iii) accuracy assessment. The multispectral bands were stacked with the calculated Normalized Difference Vegetation Index (NDVI) and the LiDAR-based attributes height, intensity, and number of returns converted into two-dimensional (2D) images. Then, a hyper-parameter analysis was performed to investigate the impact on the classification accuracy and training time of the U-Net architecture by varying the input tile size and the patch size for prediction, including the learning rate and algorithm optimizer. Finally, comparative experiments were conducted by altering the input data type to test our hypothesis, and the CNN model performance was analyzed by using accuracy assessment metrics and visually comparing the segmentation maps. The findings of this investigation showed that fusing multispectral and LiDAR data improves the classification accuracy of the U-Net, as it yielded the highest overall accuracy of 79.34% and a kappa coefficient of 0.6966, compared to using multispectral (OA: 76.03%; K: 0.6538) or LiDAR (OA: 37.79%; K: 0.0840) data separately. Although some confusion still exists among the seven land cover classes observed, the U-Net delivered a detailed and quite accurate segmentation map. [ABSTRACT FROM AUTHOR]

Skin cancer is a prevalent and perilous form of cancer and presents significant diagnostic challenges due to its high costs, dependence on medical experts, and time-consuming procedures. The existing diagnostic process is inefficient and expensive, requiring extensive medical expertise and time. To tackle these issues, researchers have explored the application of artificial intelligence (AI) tools, particularly machine learning techniques such as shallow and deep learning, to enhance the diagnostic process for skin cancer. These tools employ computer algorithms and deep neural networks to identify and categorize skin cancer. However, accurately distinguishing between skin cancer and benign tumors remains challenging, necessitating the extraction of pertinent features from image data for classification. This study addresses these challenges by employing Principal Component Analysis (PCA), a dimensionality-reduction approach, to extract relevant features from skin images. Additionally, accurately classifying skin images into malignant and benign categories presents another obstacle. To improve accuracy, the AdaBoost algorithm is utilized, which amalgamates weak classification models into a robust classifier with high accuracy. This research introduces a novel approach to skin cancer diagnosis by integrating Principal Component Analysis (PCA), AdaBoost, and EfficientNet B0, leveraging artificial intelligence (AI) tools. The novelty lies in the combination of these techniques to develop a robust and accurate system for skin cancer classification. The advantage of this approach is its ability to significantly reduce costs, minimize reliance on medical experts, and expedite the diagnostic process. The developed model achieved an accuracy of 93.00% using the DermIS dataset and demonstrated excellent precision, recall, and F1-score values, confirming its ability to correctly classify skin lesions as malignant or benign. Additionally, the model achieved an accuracy of 91.00% using the ISIC dataset, which is widely recognized for its comprehensive collection of annotated dermoscopic images, providing a robust foundation for training and validation. These advancements have the potential to significantly enhance the efficiency and accuracy of skin cancer diagnosis and classification. Ultimately, the integration of AI tools and techniques in skin cancer diagnosis can lead to cost reduction and improved patient outcomes, benefiting both patients and healthcare providers. [ABSTRACT FROM AUTHOR]

Sea level rise (SLR) attributed to the melting of ice caps and thermal expansion of seawater is of great global significance to vast populations of people residing along the world's coastlines. The extent of SLR's impact on physical coastal areas is determined by multiple factors such as geographical location, coastal structure, wetland vegetation and related oceanic changes. For coastal communities at risk of inundation and coastal erosion due to SLR, the modelling and projection of future sea levels can provide the information necessary to prepare and adapt to gradual sea level rise over several years. In the following study, a new model for predicting future sea levels is presented, which focusses on two tide gauge locations (Darwin and Milner Bay) in the Northern Territory (NT), Australia. Historical data from the Australian Bureau of Meteorology (BOM) from 1990 to 2022 are used for data training and prediction using artificial intelligence models and computation of mean sea level (MSL) linear projection. The study employs a new double data decomposition approach using Multivariate Variational Mode Decomposition (MVMD) and Successive Variational Mode Decomposition (SVMD) with dimensionality reduction techniques of Principal Component Analysis (PCA) for data modelling using four artificial intelligence models (Support Vector Regression (SVR), Adaptive Boosting Regressor (AdaBoost), Multilayer Perceptron (MLP), and Convolutional Neural Network–Bidirectional Gated Recurrent Unit (CNN-BiGRU). It proposes a deep learning hybrid CNN-BiGRU model for sea level prediction, which is benchmarked by SVR, AdaBoost, and MLP. MVMD-SVMD-CNN-BiGRU hybrid models achieved the highest performance values of 0.9979 (d), 0.996 (NS), 0.9409 (L); and 0.998 (d), 0.9959 (NS), 0.9413 (L) for Milner Bay and Darwin, respectively. It also attained the lowest error values of 0.1016 (RMSE), 0.0782 (MABE), 2.3699 (RRMSE), and 2.4123 (MAPE) for Darwin and 0.0248 (RMSE), 0.0189 (MABE), 1.9901 (RRMSE), and 1.7486 (MAPE) for Milner Bay. The mean sea level (MSL) trend analysis showed a rise of 6.1 ± 1.1 mm and 5.6 ± 1.5 mm for Darwin and Milner Bay, respectively, from 1990 to 2022. [ABSTRACT FROM AUTHOR]

Bias in facial recognition systems often results in unequal performance across demographic groups. This study addresses this by investigating how dataset composition affects the performance and bias of age estimation models across ethnicities. We fine-tuned pre-trained Convolutional Neural Networks (CNNs) like VGG19 on the diverse UTKFace dataset (23,705 samples: 10,078 White, 4526 Black, 3434 Asian) and APPA-REAL (7691 samples: 6686 White, 231 Black, 674 Asian). Our approach involved adjusting dataset compositions by oversampling minority groups or reducing samples from overrepresented groups to mitigate bias. We conducted experiments to identify the optimal dataset composition that minimizes performance disparities among ethnic groups. The primary performance metric was Mean Absolute Error (MAE), measuring the average magnitude of prediction errors. We also analyzed the standard deviation of MAE across ethnic groups to assess performance consistency and equity. Our findings reveal that simple oversampling of minority groups does not ensure equitable performance. Instead, systematic adjustments, including reducing samples from overrepresented groups, led to more balanced performance and lower MAE standard deviations across ethnicities. These insights highlight the importance of tailored dataset adjustments and suggest exploring advanced data processing methods and algorithmic tweaks to enhance fairness and accuracy in facial recognition technologies. [ABSTRACT FROM AUTHOR]

As crime rates continue to rise, law enforcement agencies face the challenge of dealing with increased demand for their services. However, with the use of intelligence, there is an opportunity to enhance prediction and prevention efforts. A study has introduced an approach that optimizes neural networks for predicting crime trends over time and space. The methodology involves utilizing Recurrent Neural Networks to analyse behaviour patterns, Deep Convolutional Neural Networks for understanding traits, and Cross Deep Learning for integrating facial recognition capabilities. The model is trained on crime data from a specific location, along with various spatial, temporal, and demographic factors. Moreover, extracting insights from these datasets through data mining is critical. Thorough testing is conducted to improve the accuracy of identifying behaviours, enabling targeted crime prevention strategies. The advanced models have demonstrated success in forecasting crime hotspots with accuracy, which can help in allocating resources effectively. By implementing surveillance and analysis in cities, not only are security measures enhanced, but the burden on law enforcement is also reduced. This allows them to focus on critical tasks. This approach underscores the potential of optimized deep learning algorithms and neural networks in predicting activities based on data analysis. These AI-driven tools offer intelligence to support resource allocation as police departments adopt cutting-edge practices. The proposed framework can be expanded to regions to address emerging security challenges. In conclusion, this groundbreaking approach highlights the importance of using advanced technologies in crime prevention, enabling law enforcement agencies to serve communities more effectively. [ABSTRACT FROM AUTHOR]