Search

Today, wind and solar energy sources have opened their place in the power system due to their environmental appeal. With the presence of these renewable energy sources (RESs), forecasting net load (NL) and its ramp are of huge importance owing their fundamental role in evaluating and improving the flexibility of the power grid. However, it is challenging to analysis flexibility and NL ramps using NL forecasts due to the uncertainty of RESs. Furthermore, the surge in the number of electric vehicles (EVs) as notable electricity consumers, and how their charging strategy can affect the NL curve are matters of concern. Most NL forecasting and flexibility analyses overlook the concurrent integration of wind and solar sources. Moreover, the impact of controlled EVs charging on grid flexibility remains largely unexplored. Existing studies also fail to assess the proper combination of wind and solar energy sources for future flexibility needs. This paper presents a framework for forecasting NL and detecting ramps with the simultaneous presence of EVs and RESs to investigate the impact of controlled charging of EVs on flexibility. Investigating charging of EVs as a solution is delved deeply in this paper to reduce NL ramps and improve flexibility. The numerical results show that controlled EVs charging may lead to 3% ramp reduction and 52% increment in the system flexibility. This paper also proposes to forecast and find the proper combination of RESs for the next years, leading to a 75% increment in the system flexibility compared to the business as usual RESs penetration. [Display omitted] [ABSTRACT FROM AUTHOR]

With the rapid development of blockchain and digital currencies, cybercrimes targeting blockchain transactions are proliferating. Various abnormal transaction behaviors caused by malicious nodes and malicious transactions have emerged in blockchain systems. Therefore, blockchain transaction security research is a hot issue in blockchain research. Deep learning has made remarkable achievements in image and audio. However, deep learning is less effective when processing text-like data than tree-based models. Therefore, for the highly unbalanced textual financial dataset, this paper designs a deep learning anomaly detection method called 1D SA-Inception by combining CNN with Transformer structure on the basis of traditional one-dimensional CNN (1D-CNN), introduces the Inception structure to learn the information of different scales of tabular data, and introduces the Transformer (self-attention) self-attention mechanism for the research of blockchain anomaly transaction detection technology, effectively improving the prediction effect. Finally, by conducting experiments on the extreme class imbalance dataset, the results show that the 1D SA-Inception designed in this paper outperforms the traditional CNN with AUC, G-mean, and F1 indexes of 91.05 %, 84.95 %, and 83.52 %, respectively, on the textual dataset. This method can more efficiently achieve the detection of abnormal transaction behavior, safeguard consumer rights and interests, and ensure the safe development of the financial sector. [ABSTRACT FROM AUTHOR]

Accurate diagnosis of Lithium-ion batteries (Li-ion batteries) degradation plays a critical role in improving the maintenance of energy storage technology. This paper presents a method based on a novel deep network model combined with a data transformation technique to diagnose Li-ion battery degradation modes. Different from conventional studies based on specific experimental and numerical methods to estimate and predict the degradation, the proposed method is based on data-driven approach, by leveraging datasets consisting of voltage/capacity curves, these were converted into incremental capacity (IC) curves and then transformed into images using the gramian angular summation field (GASF) technique. The study adopted two models: Inception-v3 and the proposed model, both underwent fine-tuning and a subsequent transfer learning process. Degradation modes, namely loss of lithium inventory (LLI) and the loss of active materials in both the positive (LAMPE) and negative electrodes (LAMNE), were diagnosed in relation to IC curves. Finally, the model was tested using two different datasets, and the results showed that the proposed method achieved high performance, especially across three Li-ion batteries, three degradation modes, three cells, and various cycles (totaling 378 cases) the proposed method outperformed in 233 cases, thereby outperforming other methods in comparison. Our method provides a flexible data-driven approach that accurately predicts various degradation modes across different cell chemistries throughout their lifespan. [ABSTRACT FROM AUTHOR]

Distributed generation sources (e.g. photovoltaic, wind power) are raised to fit the energy demands recently in proficient way. Further, adoption of renewable energy sources to the smart grid causes severe issues in quality of power signals. These problems can be overcome with correct identification and classification of power quality disturbances (PQDs) efficiently by the both producer and consumer. In this work, a new method for classification of PQDs based on signal processing technique and deep convolutional neural network (DCNN) is proposed. The proposed method compress the raw data of PQDs by using discrete wavelet transform (DWT) and multi-resolution analysis (MRA). Moreover, CNN has the ability to automatic feature selection and classification in a single stage instead of conventional methods i.e. artificial neural networks (ANNs). In this paper, DWT 1D-CNN is designed to extract and optimize multi-scale characteristics of disturbances. Further, 1-D convolutional, pooling and batch-normalization layers are added to facilitate normalize data, better generalization ability and, effective classification process against noisy data. A typical 11Kv distribution network is simulated in MATLAB to prove the validity of the proposed method in an essential part of smart grid. [ABSTRACT FROM AUTHOR]

The optimal integration of Renewable Energy Sources (RESs) into distribution networks is crucial for sustainable energy generation. However, existing approaches utilized for optimal RES integration, including metaheuristic and analytical methods, exhibit limitations. Analytical methods are simple to implement, computationally efficient, and able to model power system mechanisms, yet they face scalability issues, making them inefficient for large-scale distribution systems. On the other hand, metaheuristic approaches, while flexible for complex problems, suffer from premature convergence problems, limiting their application in real distribution networks. To address these issues, this paper introduces a novel hybrid model that combines the strengths of analytical and metaheuristic techniques. The proposed model employs a deep learning-based method called bi-directional long short-term memory (B-LSTM) network is utilized to handle the uncertainties associated with the Photovoltaic (PV) and Wind Turbine (WT) generation and load demand. Subsequently, a distinctive analytical multi-objective index-based procedure is developed to minimize the real and reactive energy loss and voltage deviation by analyzing the branch current change caused by the RES penetration in distribution networks. The optimization problem is then solved using the binary particle swarm optimization (BPSO) technique. Importantly, the presented method considers the long-term nature of optimal RES integration and reflects the seasonal impacts of the load and RES output powers. The proposed approach is applied to the IEEE 33-bus test system. The simulation results show that the introduced hybrid model, by addressing the limitations of existing approaches, offers a promising solution for efficient and effective RES integration in distribution systems while requiring low computational effort, thereby advancing the transition towards sustainable energy infrastructure. [ABSTRACT FROM AUTHOR]

In today's digital age, the proliferation of network-connected devices has triggered a surge in cyberattacks. Distributed Denial-of-Service (DDoS) attacks pose a particularly formidable challenge to network security by disrupting access to vital services. While numerous researchers have proposed DDoS detection methods utilizing machine learning and deep learning techniques, developing a robust and reliable DDoS intrusion detection system remains challenging. This challenge is exacerbated by issues such as highly imbalanced data, multi-classification, and computational complexity. This paper proposes an innovative feature selection approach to create a robust intrusion detection system capable of detecting and classifying recent common DDoS attack types. We evaluate the performance of our model on the CICDDoS2019 benchmark dataset. Our experimental results demonstrate that our proposed model outperforms existing methods, achieving a detection accuracy of 96.82%, a recall of 96.82%, a precision of 96.76%, and an F1 score of 96.50%. Additionally, our model exhibits faster prediction times, with the ability to predict an attack in just 0.189 ms. Notably, our approach, combined with preprocessing and feature selection techniques, outperforms previous works and baseline models in DDoS attack classification. [Display omitted] • An innovative hybrid feature extraction method for training of DDoS attack detection model. • We propose a robust CNN model integrated with an Inception mechanism designed to classify various types of DDoS attacks. • Evaluated the network performance of the proposed system in terms of various parameters and compared the results with existing studies and baseline models. • The proposed model outperforms more than 96% in all performance metrics (Precision, Recall, Accuracy, and F1 Score). • The proposed model achieves the fastest prediction time of 0.189 ms among existing models in the literature. [ABSTRACT FROM AUTHOR]

• Manual liver segmentation from CT scans is error-prone and time-intensive, necessitating the need for automated methods. • A novel automated framework, GWO-SwinUNet combines Gray Wolf optimized Swin Transformer with UNet architecture for liver segmentation. • GWO-SwinUNet achieves superior segmentation accuracy with remarkable Dice and Jaccard coefficients. • The hybrid approach enhances fine-grained detail extraction while ensuring robustness and convergence speed. Liver segmentation in medical images, particularly CT scans, is crucial for diagnosing and treating liver diseases. However, manual segmentation by radiologists is time-consuming and prone to errors, necessitating automated solutions. Although deep learning methods have shown promise, issues such as model overfitting, convergence, and sensitivity to noise persist. In response, this paper introduces GWO-SwinUNet, a novel liver segmentation framework that combines the Gray Wolf optimized Swin Transformer with the UNet architecture. By leveraging this hybrid approach, the method enhances fine-grained details while maintaining robustness and convergence speed. Extensive experimentation on diverse datasets and comparisons with state-of-the-art methods demonstrate the superior performance of GWO-SwinUNet, achieving a Dice coefficient of 0.988 and a Jaccard coefficient of 0.979. A thorough ablation study further validates the efficacy of the proposed strategy. In summary, GWO-SwinUNet represents a significant advancement in liver segmentation, harnessing the strengths of transformer-based architectures and optimization techniques to enhance accuracy and efficiency in medical image analysis. [Display omitted] [ABSTRACT FROM AUTHOR]

Mobile devices have become an essential element in our day-to-day lives. The chances of mobile attacks are rapidly increasing with the growing use of mobile devices. Exploiting vulnerabilities from devices as well as stealing personal information, are the principal targets of the attackers. Researchers are also developing various techniques for detecting and analyzing mobile malware to overcome these issues. As new malware gets introduced frequently by malware developers, it is very challenging to come up with comprehensive algorithms to detect this malware. There are many machine-learning and deep-learning algorithms have been developed by researchers. The accuracy of these models largely depends on the size and quality of the training dataset. Training the model with a diversified dataset is necessary to predict new malware accurately. However, this training process may raise the issue of privacy loss due to the disclosure of sensitive information of the users. Researchers have proposed various techniques to mitigate this issue, such as differential privacy, homomorphic encryption, and federated learning. This survey paper explores the significance of applying federated learning to the mobile operating systems, contrasting traditional machine learning and deep learning approaches for mobile malware detection. We delve into the unique challenges and opportunities of the architecture of in-built mobile operating systems and their implications for user privacy and security. Moreover, we assess the risks associated with federated learning in real-life applications and recommend strategies for developing a secure federated learning framework in the domain of mobile malware detection. [ABSTRACT FROM AUTHOR]

• A comprehensive and up-to-date review of deep learning-based encryption for digital images is presented. • A detail overview of image encryption using deep learning models along with recent applications are discussed. • Comparative analysis of the reviewed methods and their limitations are discussed to guide future directions. With the innovation of high-throughput big data storage technology and the construction of digital information systems in media applications, images are the common and most popular media carriers and are generated and transmitted in huge quantities. Due to the significant advantages and high value of digital image content, there is an urgent demand to protect their privacy and prevent leakage. In the past few years, deep learning models have brought more options for image encryption algorithm design. This paper offers a comprehensive survey of recent digital image encryption using deep learning models. First, a strong motivation for image encryption using deep learning models along with recent applications is introduced. Next, various state-of-the-art deep learning-based encryption techniques are discussed. The technical summary of different popular techniques is then given in tabular form. Lastly, by deeply investigating existing deep learning-based encryption, we establish some important research challenges and possible solutions, including standard security metrics. [Display omitted] [ABSTRACT FROM AUTHOR]

Accurate short-term wind power prediction is crucial for the efficient and safe operation of wind power systems. To enhance the accuracy of short-term wind power prediction, this paper proposes a hybrid short-term wind power prediction model, IDBO-VMD-TCN-GRU-Attention, based on the Improved Dung Beetle Optimizer algorithm (IDBO), Variational Modal Decomposition (VMD), Temporal Convolutional Network (TCN), Gated Recurrent Unit (GRU), and Attention Mechanism. The IDBO is used to optimize the parameters of the VMD. Then the optimized IDBO-VMD is used to decompose the original data into modal components with lower volatility to fully extract the data features. These modal components are inputted into the TCN-GRU-Attention to predict the short-term wind power and obtain the final prediction value. The model is tested and validated using real data from four different months and compared against 11 other models. Results demonstrate that our proposed model significantly reduces the Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE) by an average of at least 22.86 %, 18.82 %, and 19.99 %, respectively, across the four different months. This indicates that the proposed model provides a superior fit to the data, offers higher prediction accuracy, and holds significant practical value. [ABSTRACT FROM AUTHOR]

Localization models, particularly for indoor applications like autonomous vehicles and smart manufacturing in Industry 4.0, have emerged as a crucial research area because of the growing need for precise positioning. In this context, distributed cell-free MIMO have demonstrated capabilities for a higher fine-grained resolution and accuracy. This paper explores different approaches to address challenges related to positioning and corresponding data transmission and model compression problems in cell-free massive MIMO systems. We present a novel deep-learning powered CSI compression model for reducing the signalization data associated with the localization task, minimizing the fronthaul capacity requirement by at least a factor of 2, without a significant loss of accuracy. In our research, we also focus on the challenges of deploying deep end-to-end neural networks on resource-constrained embedded platforms. Two approaches are proposed: model modification and magnitude-based pruning. Our results show that the pruned end-to-end localization model has a smaller size by 36% at the cost of an increase on the localization error by a factor of 1.5 (precision is reduced from approximately 5 mm to 7.8 mm). [ABSTRACT FROM AUTHOR]

Many IoT devices are presently in use without sufficient security measures. The vulnerability of these devices to malware highlights the necessity for effective methods to identify malicious activity while protecting data privacy. Federated Learning is a collaborative learning technique that can create a fair model while keeping the local data private. To this end, this paper presents the SIM-FED, a novel model for detecting malware in IoT devices. This model utilizes both deep learning and federated learning algorithms. The SIM-FED offers a privacy-preserving solution for IoT malware detection, without sharing data, enhancing security in distributed environments. The model utilizes a lightweight one-dimensional CNN with optimized hyperparameters, reducing preprocessing, and computational overhead. Various federated aggregation strategies are evaluated, and the FedAvg strategy is selected to integrate the results of local models in the proposed model. Moreover, the model's resilience against white-box and black box cyber-attacks is assessed, revealing minimal performance degradation and highlighting its robustness, which is often overlooked in previous studies. The performance of the SIM-FED is assessed by conducting a series of experiments using the IoT-23 dataset. The results show that the SIM-FED outperforms other deep learning and federated learning models in terms of all evaluation metrics, achieving a remarkable 99.52 % accuracy. [ABSTRACT FROM AUTHOR]

• A comprehensive overview of recent advances in visual odometry (VO), focusing on deep learning models. • An in-depth model comparison, highlighting explainability and interpretability aspects. • A summarization of research gaps and challenges in the context of explainable AI (XAI) in visual odometry. • A discussion on advances in XAI techniques for visual odometry, allowing transparent models. • An overview of practical implications of visual odometry advancements in real-time applications. Visual Odometry (VO) is a crucial process for estimating camera motion in real-time based on visual information captured. The emergence of deep learning has significantly transformed VO and Explainable Artificial Intelligence (XAI) in Deep Vision-Based Odometry. This survey paper explores the latest advancements in VO facilitated by deep learning methods, focusing on explainability and interpretability. It provides an overview of state-of-the-art deep learning techniques and dissects each model into its elemental building blocks to understand their explainable and interpretable aspects. The survey also highlights research gaps in optical flow robustness, occlusion and dynamic objects, real-time processing, drift correction, semantic awareness, and sensor integration. The aim is to catalyze future innovations in deep learning-based VO and stimulate dialogue about potential directions for the next wave of research, emphasizing explainability and interpretability as integral components of advanced systems. [Display omitted] [ABSTRACT FROM AUTHOR]

This paper presents an effective and accurate method to analyze the network traffic risk of enterprise public cloud financial system by using deep learning algorithm. To build the analytical model, the collected data is preprocessed to extract relevant features that contribute to risk analysis. The pre-processed data is used to train the deep learning model. The model is designed to learn patterns and relationships in network traffic data. The network traffic risk analysis module is constructed by combining the deep learning model with the system architecture of the enterprise public cloud financial system. Through a trained deep learning model, the module is able to accurately identify abnormal behavior in network traffic and provide corresponding risk assessments. The results show that the proposed system module can accurately identify abnormal behavior in network traffic and provide risk assessment. [ABSTRACT FROM AUTHOR]

A modern smart grid tends to have increasingly various uncertain renewable generations. Due to different geographical areas and network topology constraints, operations of a smart grid become complicated and challenging. Moreover, using existing methods, power flow estimation in real-time could be time-consuming and computationally expensive. This paper proposes an efficient deep learning approach to estimate real-time power flow solutions of the smart grid with high penetration of uncertain PV generations using synchrophasor data considering measurement noise. The performance and effectiveness of the proposed convolutional long short-term memory (ConvLSTM) with time-series cross-validation technique are examined using synchrophasor data with Gaussian noise in the IEEE 39 bus test system. The proposed ConvLSTM approach shows better robust performance than weighted least square (WLS) state estimation and long short-term memory (LSTM) approaches. In addition, state measurements and confidence intervals are employed to confirm the accuracy of estimated real-time power flow results. The accurate real-time power flow estimation is crucial to determining dynamic available transfer capability (ATC) results and efficient operations of smart grids. [Display omitted] • The proposed deep learning ConvLSTM approach provides accurate real-time power flow results. • The real-time power flow estimation employs synchrophasor data with measurement noise. • The proposed approach is efficient and effective for a power system with uncertain PV. • The accurate results are confirmed by means of state measurements and confidence intervals. [ABSTRACT FROM AUTHOR]

Advancements in deep learning for speaker identification are constrained by the limited availability of data, especially in law enforcement applications. This has led to the emergence of few-shot speaker identification, a technique that classifies unseen test samples with the help of a few support samples. Despite several attempts to advance few-shot speaker identification, significant challenges persist, including the extraction of robust speaker embeddings, the problem of overfitting, and the issue of prototype shift error. This paper proposes a few-shot speaker identification system employing a novel architecture called the Channel Attention-based Convolutional Recurrent Neural Network (CACRN-Net) with three-dimensional (3D) log Mel spectrogram inputs to mitigate overfitting and enhance the accuracy of speaker embeddings. Furthermore, a self-attention mechanism alleviates prototype shift errors caused by noisy data. The proposed framework is compared to existing methods using VCTK and Voxceleb1 speech corpora through 5-way, 5-shot learning experiments. To assess the performance of the framework in speech variability conditions, we utilized the IIT Guwahati (IITG) multi-variability (MV) speech database. The proposed approach outperforms state-of-the-art techniques, achieving a substantial enhancement in speaker identification with a 2.73 % accuracy improvement on the VCTK database and a 2.3 % improvement on Voxceleb1. [ABSTRACT FROM AUTHOR]

The rapid growth of online music education has led to increased security risks from cyber intrusions. This paper proposes public cloud networks oriented deep neural networks for effective intrusion detection in online music education environments. Specifically, a novel intrusion detection framework is developed, comprising fuzzy logic based feature selection, chronological salp swarm algorithm optimized deep belief networks, and gated recurrent unit integrated convolutional neural networks. Detailed methodologies are presented for the fuzzy logic system, chronological salp optimization, deep belief network architecture, and convolutional neural networks. Comprehensive experiments are conducted on the NSL-KDD and CICIDS2017 datasets. Various deep neural networks are evaluated and compared, including multi-layer perceptrons, convolutional neural networks, deep belief networks, recurrent neural networks, and the proposed models. Experimental results demonstrate that the proposed fuzzy feature selection and chronological salp swarm algorithm optimized deep belief network achieves a test accuracy of 97.33 %, outperforming other peer models. The gated recurrent unit integrated convolutional neural network obtains a test accuracy of 98.46 %, superior to state-of-the-art methods. While the experiments on the newly created dataset for intrusion detection in cloud-based online music education demonstrate that the proposed models outperform the benchmarks. The experiments verify the effectiveness of the proposed deep learning frameworks for intrusion detection in online music education cloud networks. [ABSTRACT FROM AUTHOR]

• A method of exchange rate market trend prediction based on sentiment analysis. • Forecast the exchange rate market trend by combining CNN with LSTM. • The "emotional" factors in the text data of Weibo are excavated. • It can still perform stably on small data sets. Exchange rate market is an important part of the global economy to reflect national economic development.With the rise of the Internet, more and more people choose to express their opinions on the Internet. Therefore, we believe that the comments in the online text will also reflect the trend of the exchange rate market. In this paper, a hybrid model of exchange rate market trend prediction based on sentiment analysis is proposed. The sentiment analysis module is constructed by combining Word2vec with Long Short-Term Memoryto extract "emotion" factors in text data and improve the performance of the model.After data cleaning, the model generates word vector and conducts sentiment analysis to get three emotional classifications of "positive", "ordinary" and "negative" by crawling the contents of "exchange rate" entries on "Weibo", and obtains the emotional weight of each day.Then, the emotional weights of about 3 months were randomly inserted into the historical exchange rate data of 3-4 years to conduct the mixed training combining Convolutional Neural Network and LSTM.In this paper, the historical exchange rate data of RMB against US dollar, Japanese yen, Euro and Australian dollar in 3-4 years are used as experimental samples. Experimental results show that compared with some traditional models, the method proposed in this paper is used to conduct comparative analysis by Mean absolute error, Mean squared error, Mean absolute percentage error and R2_Score after inserting small emotional data. The results show that our proposed method outperforms all comparison models in all indicators. Taking the trend prediction of the Australian dollar/RMB exchange rate as an example, our method outperforms the comparison models by an average of about 25.32% in the MAE results, 38.51% in the MSE results, and 2.76% in the MAPE results, R2_Score The average result of Score is 6.16% better than the comparison models, indicating a certain degree of robustness. In order to verify the effectiveness of the emotional weights we extracted in predicting exchange rate trends, we extracted the proposed method from the ''emotional analysis'' module and conducted a self comparison. The results showed that the results of the module with the addition of ''emotional analysis'' were significantly better than those without the addition of ''emotional analysis" module. [Display omitted] [ABSTRACT FROM AUTHOR]

Non-contact facial video-based physiological signal measurement using remote photoplethysmography (rPPG) has shown great potential in many applications (e.g. telemedicine, driver fatigue detection, sleep detection) and has achieved reliable results in the controlled environment. With the development of deep learning, more and more methods have been proposed. However, these methods were usually implemented based on model variations that have achieved better results in computer vision tasks, and finally chose average pooling layer or one-dimensional convolution layer directly reduce dimension. Such these operations ignored the fact that the physiological signals of one individual are consistent in different facial regions. In this paper, we propose a novel remote photoplethysmography framework that utilizes graph neural network to extract physiological signal, named GraphPhys. And we design the Average Relative GraphConv based on the fact that the physiological signals of one individual are always consistent in different facial regions for the remote physiological signal measurement task. We apply GraphPhys to two different architectures for rPPG methods, and the methods based on GraphPhys significantly outperform the original methods on the VIPL-HR (Niu et al., 2019) dataset. To the best of our knowledge, this is the first work to utilize graph neural network to aggregate physiological signals. The code is available at https://github.com/Xiong-JiaHao/GraphPhys. [ABSTRACT FROM AUTHOR]

Traditional network traffic analysis methods fall short in addressing the requirements of complex network environments. Therefore, the introduction of advanced technologies, such as deep learning, is necessary to enhance the accuracy and efficiency of network traffic analysis. This paper designs a risk analysis method of enterprise cloud financial system network traffic based on deep learning to improve the level and effect of network security assurance. A deep learning based traffic analysis framework has been constructed, which includes several main steps such as data preprocessing, feature extraction, model training, and result evaluation. The test results on the experimental data set show that this method can identify network attacks and abnormal traffic more quickly and accurately than the traditional network traffic analysis methods, and has better practicability and application prospects. [ABSTRACT FROM AUTHOR]

Distributed Denial of Service (DDoS) attacks are common and increasing in frequency. It renders the service inaccessible to legitimate users and degrades network performance. A complex structural environment called a Cyber–Physical System (CPS) is created by combining computation, connectivity, and physical parameters. Software-Defined Networking (SDN) is an emerging architecture that separates the data plane from the network plane. A central logic control resides in the control plane, making SDN vulnerable to DDoS attacks. SDN design ideas are broadened and applied to create software-defined Cyber–Physical Systems. Deep learning powers many artificial intelligence apps and services, enhancing automation by performing cognitive tasks without human intervention. It can perform feature extraction and classification on both small and large datasets. This paper presents a variety of Deep Learning models for efficiently detecting DDoS attacks in the SD-CPS framework through a scalable and adaptable SDN-based architecture. We could determine which Deep Learning techniques work best under various attack scenarios by examining multiple Deep Learning techniques. The Deep Learning models performed above 99% accuracy in classifying binary and multiclass data over unknown traffic when tested on two recent security datasets, the SDN-specific dataset and the CICDDoS2019 dataset. [Display omitted] • The paper utilizes two datasets, one tailored to SDNs and the CICDDoS2019 dataset. • Three distinct deep learning approaches identify binary and multiclass attacks. • The binary accuracy is over 99% for the datasets, which is better than prior studies. • Multiclass detects attack type with accuracy greater than 99±% for both datasets. • The individual accuracy for type of attack ranges between 98%–99% in most cases. [ABSTRACT FROM AUTHOR]

• We propose a personalized course recommendation method based on BERT model, which integrates deep learning and big data technology based on MOOC system. • We design a domain feature difference learning strategy, which is able to learn the difference features between course contents to better extract information in the text and improve the recommendation performance of the model. • We experimentally demonstrate the effectiveness of the method proposed in this paper on the open MoocCube datasets. Finding the courses that users are interested in quickly in the massive data can make a very important contribution to the accurate dissemination of knowledge. In this paper, we integrate the deep learning and big data technology to investigate a personalized recommendation method based on Massive Open Online Course (MOOC) system. Based on the Bidirectional Encoder Representations from Transformers (BERT) model, we propose some corresponding strategies to improve the accuracy of the recommendation system. First, we introduce the acquisition and preprocessing of the open dataset. Second, we design a recommendation model framework by taking advantage of the BERT model and incorporating a self-attention mechanism. Finally, to obtain deep feature information between course texts, we design a domain feature difference learning strategy to improve the model's recommendation performance. The results of our experiments prove that the proposed model in this paper performs good recommendation results compared with other methods. [Display omitted] [ABSTRACT FROM AUTHOR]

Smart process repositories have been developed to effectively manage large collections of business process models. Several key features of such repositories require accurate matching of corresponding activities between a pair of process models. However, the existing matching techniques have not yet achieved the desired accuracy, which impedes the effectiveness of process repositories. To that end, this paper proposes a word-embedding-based approach that significantly improves the accuracy of matching. For a comprehensive evaluation of the proposed approach, we have performed experiments using three state-of-the-art word embeddings, two syntactic measures, six semantic measures, and four datasets. The results show that the use of word embeddings outperforms all the syntactic as well as the semantic similarity measures. Moreover, the use of fastText-based embeddings in our proposed technique achieves the highest F1 score, compared to both Word2vec- and GloVe-based embeddings. [ABSTRACT FROM AUTHOR]

• License plate detection for multi-national vehicles in a multi-lane scenario. • Ability to deal with multi-national license plates of various colors and styles. • Capability to handle illumination variance. • Performance evaluation on datasets of China, Thailand, Pakistan, USA, UAE and Netherlands. • Superior performance as compared to conventional and deep learning approaches. Only a few methods in literature are effective for multi-national license plate detection in a multi-lane scenario. These methods are prone to illumination variance, complex background and weak-edged license plates. In this paper, we propose a novel illumination invariant method to handle multi-national vehicle license plates of different colors and styles. Red corona is initially used to detect the tail-lights of vehicles to establish region-of-interest as the license plates are in a vicinity of its tail-lights. The vertical edges within each region-of-interest are obtained using a unique approach that preserve license plate edges for improved performance. Heuristic energy map is then used to distinguish the license plate area. To validate the detected regions, high-level features extracted from AlexNet Convolutional Neural Network are used. Extensive experiments on the license plates of six countries show that the proposed approach not only ensures real-time performance, but also outperforms the conventional and deep-learning methods. [ABSTRACT FROM AUTHOR]

• Interpretable DL Model: Our unique deep learning model interprets modulation mechanisms, efficiently extracting time and phase features with minimal resources. • Efficient Performance: Our approach achieves top performance on RML2016 and RML2018 datasets with fewer model parameters. • Fastest Embedded Speed: Outperforming 8 DL-based methods, our approach demonstrates the fastest speed on Jetson Nano and Raspberry Pi systems. Automatic Modulation Classification (AMC) constitutes a fundamental technology for enabling automatic demodulation in Cognitive Communication Systems (CCS). Due to the size, weight, and power (SWaP) constraints of embedded computers employed in CCS, there are limited computational and memory resources. While deep neural networks possess strong feature representation and high accuracy recognition capabilities, they usually come with a high number of network parameters and high computational complexity, thereby reducing the real-time processing ability of CCS. Therefore, neural network structures intended for CCS must be lightweight and computationally efficient. In this paper, we propose a high-performance and resource-friendly network model based on an analysis of the modulation mechanism of communication signals. The network extracts phase features and short-time features sequentially using directional convolutional filters. Long short-term memory (LSTM) units are then used to extract long-term features, and only one fully connected layer is used for classification. Experiments with a standard dataset consisting of 11 communication modulation types demonstrate that our proposed model achieves an accuracy greater than 84.5%, even when the signal-to-noise ratio (SNR) is 0 dB, and the model has only 29187 parameters. On a Jetson Nano embedded platform, the model achieves a processing speed of up to 375366 in-phase and quadrature samples/s. Overall, the results suggest that our proposed approach is both lightweight and highly efficient, making it more suitable for CCS applications. [Display omitted] [ABSTRACT FROM AUTHOR]

White blood cells (WBCs) are crucial constituents of the blood that protect the human body against infections and viruses. The classification of WBCs in a blood smear image is used to diagnose a range of haematological disorders. The manual identification of WBCs can result in potential errors, necessitating the need for automated systems that can assist in classification. Recently, various deep learning models, such as DenseNet121, Xception, MobileNetV2, ResNet50, and VGG16, have been used to classify WBCs. However, the available classification models are black boxes because their decisions are difficult for humans to understand without further exploration. The interpretability and explainability of these models are essential, as their decisions can have severe consequences for patients. In this paper, we integrate an explainable AI (XAI) technique called local interpretable model-agnostic explanations (LIME) with the DenseNet121 classification model for WBC classification. Interpretable results would allow the users to understand and verify the model's predictions, enhancing their confidence in the automated diagnosis. • Comparison of deep learning models for leukocyte classification. • Inclusion of explainability to visually explain the final decision. • The results show high explainability of the prediction mechanism. [Display omitted] [ABSTRACT FROM AUTHOR]

The use of Neural Architecture Search (NAS) techniques is becoming more widespread. NAS methods have been essential for automating and accelerating the laborious and erroneous process of designing and tuning new Deep Learning (DL) approaches. The past few years have seen increasing number of research works on NAS and associated hyper-parameter optimization (HPO). They have arguably had the biggest influence on classification and object detection tasks, where they have produced state-of-the-art results. Despite the notable success to date, it is still difficult and not always practicable to apply NAS to real-world situations. In general, the structure of Convolution Neural Networks (CNNs) is too sophisticated to be used in platforms with limited resources, such as IoT, mobile, and embedded systems. In this paper, we revisit several state-of-the-art NAS and HPO methods from a multi-perspective view both in non-federated and distributed federated learning environments. Earlier attempts to critically analyze the components of NAS or HPO are only limited to non-federated environment, where NAS and HPO related computations need to be performed on a single CPU/ multiple GPUs and where all the data was readily accessible for performing the task of learning. In this article, we extend the discussion and critically analyze the important components of NAS and HPO methods in federated learning environment, where computations need to be performed on multiple computational units in federation and where only local data can be used by a single computational unit. Due to unavailability of entire data in federated learning environment, the earlier NAS methods (i.e., non-federated NAS methods) and the earlier HPO methods (i.e., non-federated HPO methods) cannot be used directly without modifications. We discuss these modifications made by researchers in non-federated NAS and HPO methods to be used in federated learning environment and provide discussion for future avenues. [ABSTRACT FROM AUTHOR]

Medical imaging is crucial for monitoring the progression of a diagnosis. However, poor-quality images, particularly radiographs which are commonly used for their affordability, can lack sufficient visual information and lead to lengthy and inaccurate diagnoses, reducing clinicians' ability to examine and decreasing the accuracy of their assumptions. To address this issue, this paper proposes an Xray-Net approach that utilizes self-supervised pixel-contrast stretching to improve low-contrast images. The approach consists of three main phases: (1) Pre-processing, which involves obtaining X-ray contrast-based classification through the measurement of average brightness, (2) Development of a conventional neural network model, and (3) Contrast-adaptive stretching, where the stretching control coefficients are predicted adaptively based on input image information. The proposed approach was evaluated using quantitative and qualitative experiments on three X-ray databases, and six different quality measures were considered. The results obtained by Xray-Net demonstrate superior low-contrast enhancement performance compared to several state-of-the-art methods. • Low-contrast radiography enhancement by adaptive intensity transformation. • Adaptive intensity transformation using pixel-stretching-based learning. • Deep model-based learning approach for self-supervised pixel stretch. • Evaluation of the proposed approach with SOTA, qualitatively and quantitatively. • Proposed approach achieves enhanced radiographs compared to related methods. [ABSTRACT FROM AUTHOR]

The Internet plays a crucial role in supporting global applications and businesses, but security remains a major challenge. Within the Internet, there exists a parallel network known as the Darknet, where malicious activities and traffic are present and require real-time classification. Many methods aim to classify this Darknet traffic in real-time due to its significant volume within Internet traffic. However, online Darknet traffic classification faces challenges, particularly in determining the optimal packet sampling amount for achieving a high classification rate in high-performance networks. To address this, our paper presents a novel approach that combines Convolutional Neural Network (CNN) and Reinforcement Learning (RL) techniques for intelligent and adaptive packet sampling rates in high-performance network interfaces. This method reduces overhead on monitored entities, especially in high-speed networks with a high bit rate. Our findings demonstrate a TOR traffic prediction accuracy of 99.84% and successful classification tasks in high-throughput networks using our method. • A short literature review on Darknet traffic classification approaches. • A near to real-time method for malicious traffic classification based on CNN. • An intelligent approach for sampling network packets for fast networks based on RL. • A Convolutional Neural Network (CNN) performance evaluation in the malicious traffic classification task. [ABSTRACT FROM AUTHOR]

Hyperspectral remote sensing imaging technology provides assistance in various aspects of daily life through applications such as urban building information statistics and green vegetation estimation. Ensuring the accuracy of automatic thematic information extraction under limited samples is a challenge. In this manuscript, a lightweight semantic segmentation model based on the "encoder-decoder" structure is proposed for extracting buildings from hyperspectral remote sensing images. The proposed model employs the lightweight MobileNet combined with multiscale feature fusion and a group dilated convolution for modelling both shallow and deep spatial and spectral features as the encoder and an efficient combined standardized attention mechanism for selecting the most valuable bands and local information. Extensive experiments reveal that our method produces greater accuracy than state-of-the-art lightweight models in building extraction tasks. We also demonstrated the superiority of our method for insufficient training sample sizes. When only 50% of the samples of the initial training set were used, the mean intersection over union (mIOU) reached 91.90%, 4.5% higher than that of the next best method. For training sets composed of only 16 and 8 images, the mIOU values were 89.42 and 77.11%, respectively, 13.6 and 18 percentage points higher than that of the next best method. According to the visualization of the results, the proposed method obviously outperformed the compared methods. The model proposed in this paper is suitable for accurately extracting buildings from hyperspectral images in situations involving limited training samples. [ABSTRACT FROM AUTHOR]

Despite its high computation costs, human activity recognition is one of the most widely researched areas in computer vision. Deep learning based Human activity recognition in the complex live video has been focused on in this work. Various works stated that end-to-end models are adequate for learning features of video inherently. However, it is a challenging method to capture effective features to recognize the activity. To alleviate this problem, we identify the proposals where the actual event is happening amongst the sequence of frames. In this paper, deep CNN and Stacked LSTM are used to select the deep features for action and maintain the long-term dependencies (i.e., previous information) of the action features, respectively. Also, we proposed Graph Convolutional Network (G-ConvNET), which encodes the features from the graph to label the activity. The proposed method is tested on the THUMOS14 dataset, which comprises twenty complex sports field activities. The experimental results show a significant improvement of 8.6% in modernized methods with mean Average Precision (mAP). [ABSTRACT FROM AUTHOR]

Leak detection in natural gas pipelines is an extremely important and persistent problem in the oil and gas industry. The construction of accurate physical models of pipelines is limited by the high complexity, unavailable closed-form solution, and strict experienced personnel requirements. Moreover, industrial automation has the common problems of large amount of data with little information. This paper proposes a data-driven digital twin (DT) method for leak detection as a new paradigm solution to these challenges. From the perspective of knowledge-based data-driven, a DT pipeline learning and updating scheme based on normal data directly from operational data during the entity pipelines life-cycle is proposed to enhance DT adaptability. A DT-driven leak detection method is proposed, making effective use of data interaction and fusion of DT. The effectiveness and performance of the proposed approach is illustrated by deploying the DT pipeline in a simulated leak scenario of a real running natural gas pipeline. © 2012 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of Global Science and Technology Forum Pte Ltd [ABSTRACT FROM AUTHOR]

With the rapid evolution of the Internet of Things (IoT), network advancement has significantly influenced the increasing number of devices and advanced enhancements linked with it. Indeed the increasing number prevalence and sophistication of emerging cyber-attacks have highlighted the necessity for designing robust security application. In this paper, the Remora-based Deep Maxout Network model is Proposed. Here, the input data is acquired and forwarded to the pre-processing phase, wherein the missing value imputation approach is employed for creating a complete dataset. Later, the pre-processed data is then subjected to dimension transformation from the transformed data; the Convolutional Neural Network features are extracted, followed by feature selection based on Canberra distance. Here, detection is carried out using a Deep Maxout Network whose weights and training parameters are modified using the Remora Optimization Algorithm. However, the proposed model has delivered superior results with a high testing accuracy of 0.945 [ABSTRACT FROM AUTHOR]

• Artificial intelligence and deep learning have good prospects in computer network security. • Improving computer network security detection rate. • Reducing false positives in computer network security. With the continuous progress of the information age and science and technology, how to combine complex and changeable data with self-protection ability has become an urgent problem that needs to be solved. In the computer system, the traditional technology for protection is somewhat passive. To this end, this paper analyzes computer network security monitoring under artificial intelligence (AI) deep learning technology and discusses its development trend, which monitors network security to find and address problems in time. The results show that the deep learning technology of computer network security detection can improve its security performance with high safety performance, high detection rate, and low false alarm rate. It can monitor the vulnerabilities in the computer network in time and realize effective security attack detection of the computer network. Deep learning technology for computer network security detection can improve its security performance, timely monitor vulnerabilities in computer networks, and achieve effective security attack detection on computer networks [Display omitted] [ABSTRACT FROM AUTHOR]

The accurate spatial-temporal prediction of photovoltaic (PV) power generation helps the power system dispatching department to make reasonable dispatching plans. In this paper, a robust spatial-temporal prediction model for PV power generation based on a denoising autoencoder (DAE) combining Gramian angular summation fields (GASF) and convolutional neural network (CNN) is proposed. First, the downscaling and modeling of power data use Pearson correlation to reduce inaccurate and inefficient predictions due to large-scale sparse data. Second, the model training data are encoded with noise reduction using a denoising autoencoder. Finally, the performance of the proposed model is experimentally verified. The results show that the model still performs well when the data exist in different degrees of contamination, with an average mean square error (MSE) of 71.58. Experiments based on China and the USA PV power generation datasets with mean absolute error (MAE) of 4.26 and 3.96, respectively. [ABSTRACT FROM AUTHOR]

Remaining useful life (RUL) prediction of rolling element bearings is critical to maintaining rotating machinery and lowering industrial costs. There are many RUL prediction techniques, but most of them ignore two factors that may have a significant impact on prediction accuracy. One is the detection of the first predicting time (FPT) while the other is the predictive uncertainty. This paper proposes a two-stage approach to incorporating both factors into the prediction process based on Bayesian deep learning (BDL). In stage one, the state change of the bearing is identified and the FPT is determined according to a proposed detection technique. In stage two, RUL prediction is performed according to a new BDL model, and the results provide RUL point estimates and quantification of predictive uncertainty. The proposed two-stage approach has been validated on two publicly available bearing datasets, and the experimental results have demonstrated the effectiveness of the proposed approach in detecting FPT and its superiority over competitive BDL models. • A new prediction framework has been proposed to address two issues in the process of predicting the RUL of bearings. • The two-stage approach achieves health stage division and RUL prediction of bearings by considering their degradation process. • The two-stage approach shows superior performance compared to other approaches on two separate stage tasks. [ABSTRACT FROM AUTHOR]

• The Intelligent Microbat Routing (IMR) and deep learning (DL) based Popularity Content Caching (PCC) IMR-PCC hybrid intelligent architecture for the caching and routing achieved the better QoS performances in vehicular edge networks. • This paper introduced the popularity based content features for achieving the better learning rate in the intelligent caching. • The framework utilized OMNET++ interfaced with Python3.8 platform for the implementation which results in better implementation of the model. • The paper focusses on the joint routing and caching technique for the vehicular edge and IoT networks. • IMR-PCC achieved the better Quality of Service (QoS) as well as executes cache performance which is highly suitable for the vehicular IoT networks. Vehicular networks are considered as one of the paramount components of intelligent transportation systems. With the help of new advanced technologies like Internet of things (IoT), and many wireless applications such as 5 G/6 G, these vehicular networks has changed its face of applications in terms of computations, communications and privacy. The arising vehicular applications such as crash awareness and emergency brake requires better communication, storage resources, and computation with severe execution prerequisites on response time and network bandwidth. But still, efficient data computation with an intelligent routing mechanism still remains to be darker side of the research. To achieve the improvisation in above parameters, this paper proposes the hybrid combination of the Intelligent Microbat Routing (IMR) and Deep learning based Popularity Content Caching (PCC) techniques which has been targeted for IoT based vehicular networks. This paper introduces the Reinforced Gated Recurrent Units (R-GRU) with High Speed Extreme Learning Machines for better prediction of contents' popularity that aids for better caching. The proposed IMR-PCC architecture consists of four different tasks such as popularity prediction, cache placement, data retrieval and effective routing. The extensive experimentation is carried out to evaluate the IMR-PCC by implementing the dynamic network topology with multi factors, which directly impacts on the performance of the complete system. Nearly 300,000 contents were collected randomly and the complete algorithm was implemented in the OMNET++ interfaced with Python 3.8. Finally from the experimental results, the proposed IMR-PCC architecture proved its excellence in terms of accuracy (98.6%), precision (98.6%), recall (98.6%), F1-Score (98.8%), cache hit ratios (0.25), latency (0.3), Packet Delivery Ratio (PDR) (99%) than other existing methodologies such as Q-learning, BLSTME, MPCF, ICN. So the proposed IMR-PCC achieved the better Quality of Service (QoS) as well as executes cache performance which is highly suitable for the vehicular IoT networks. [Display omitted] [ABSTRACT FROM AUTHOR]

With the boom in Internet and information technology, cyber-attacks are becoming more frequent and sophisticated, especially Advanced Persistent Threat (APT) attacks. Unlike traditional attacks, APT attacks are more targeted, stealthy, and adversarial, rendering it challenging to manually analyze threat behaviors for APT detection, attribution, and response. Therefore, the research community has focused on intelligent defense methods. Intelligent threat profiling is dedicated to analyzing APT attacks and improving defense capability with Knowledge Graph and Deep Learning methods. With this insight, this paper provides the first systematic review of intelligent threat profiling techniques for APT attacks, covering three aspects: data, methods, and applications. The contents include data processing techniques, threat modeling, representation, reasoning methods, etc. Furthermore, this paper summarizes the latest research in applications, proposes the research framework and technical architecture, and provides insights into future research trends. This paper contributes to recognizing the advantages and challenges of intelligent threat profiling. It paves the way for integrating knowledge graphs and deep learning to achieve intelligent security. [Display omitted] • The first review paper on intelligent threat profiling of Advanced Persistent Threat. • Summarizes the research findings on three aspects: data, methods and applications. • Proposes the research framework and technical architecture of intelligent threat profiling. • Analyzes the challenges and provides insights into future research trends. [ABSTRACT FROM AUTHOR]

• Dual fusion feature pyramids offer better performance. • The candidate box obtained by using CIOU is more friendly. • Transfer learning and data augmentation can improve the problem of data scarcity. Rail surface defects are serious to the quality and safety of railroad system operation. Due to the diversity and randomness of rail defects form, the detection of rail surface defects is a challenging task. Therefore, this paper proposes a new surface defect detection network based on Mask R-CNN to detect rail defects. The detection network is designed with a new feature pyramid for multi-scale fusion; a new evaluation metric complete intersection over union (CIOU) is used in the region proposal network to overcome the limitations of intersection over union (IOU) in some special cases; in the training phase, both transfer learning and data augmentation are used to solve the problem of small defective datasets. The experimental evaluation shows that the model proposed in this paper achieves 98.70% mean average precision (MAP) on the proposed dataset and can locate the defect location more accurately. [Display omitted] [ABSTRACT FROM AUTHOR]

Modern businesses benefit significantly from advances in computer vision technology, one of the important sectors of artificially intelligent and computer science research. Advanced computer vision issues like image processing, object recognition, and biometric authentication can benefit from using deep learning methods. As smart devices and facilities advance rapidly, current networks such as 4 G and the forthcoming 5 G networks may not adapt to the rapidly increasing demand. Classification of images, object classification, and facial recognition software are some of the most difficult computer vision problems that can be solved using deep learning methods. As a new paradigm for 6Core network design and analysis, artificial intelligence (AI) has recently been used. Therefore, in this paper, the 6 G wireless network is used along with Deep Learning to solve the above challenges by introducing a new methodology named Optimizing Computer Vision with AI-enabled technology (OCV-AI). This research uses deep learning – efficiency algorithms (DL-EA) for computer vision to address the issues mentioned and improve the system's outcome. Therefore, deep learning 6 G proposed frameworks (Dl-6 G) are suggested in this paper to recognize pattern recognition and intelligent management systems and provide driven methodology planned to be provisioned automatically. For Advanced analytics wise, 6 G networks can summarize the significant areas for future research and potential solutions, including image enhancement, machine vision, and access control. [ABSTRACT FROM AUTHOR]

Signature-based malware detection approaches are inadequate for detecting the increasingly intelligent and large number of malware programs emerging today. Therefore, alternative approaches are required. The effects of malware can be estimated by analyzing the opcodes in its executable files. It can then be classified into families using a long short-term memory (LSTM) network. Vectorizing opcodes and application programming interface (API) function names using one-hot encoding results in high-dimensional vectors because each case is represented using one dimension. Therefore, this paper proposes a word2vec-based LSTM method to analyze opcodes and API function names using fewer dimensions. The results of opcode and API function name classification using the proposed method and one-hot encoding were compared using the Microsoft Malware Classification Challenge dataset. The proposed method showed approximately 0.5% higher performance than the one-hot encoding-based approach. [ABSTRACT FROM AUTHOR]

Deep learning technology has been widely used in object detection. Although the deep learning technology greatly improves the accuracy of object detection, we also have the challenge of a high computational time. You Only Look Once (YOLO) is a network for object detection in images. In this paper, we propose a real-time object detection algorithm for videos based on the YOLO network. We eliminate the influence of the image background by image preprocessing, and then we train the Fast YOLO model for object detection to obtain the object information. Based on the Google Inception Net (GoogLeNet) architecture, we improve the YOLO network by using a small convolution operation to replace the original convolution operation, which can reduce the number of parameters and greatly shorten the time for object detection. Our Fast YOLO algorithm can be applied to real-time object detection in video. [ABSTRACT FROM AUTHOR]

• We proposed a 9-layer deep convolutional neural network for leaf disease identification. • We used six different types of image augmentation methods, and compared their validation performances using data augmentation and not using data augmentation. • Fine-tuning the batch size, dropout and epoch improves the performance of our model. • We validated that the proposed model performs better than transfer learning techniques. • Our model produced an average testing accuracy of 96.46%, which was better than the state-of-the-art machine learning approaches. In this paper, we proposed a novel plant leaf disease identification model based on a deep convolutional neural network (Deep CNN). The Deep CNN model is trained using an open dataset with 39 different classes of plant leaves and background images. Six types of data augmentation methods were used: image flipping, gamma correction, noise injection, principal component analysis (PCA) colour augmentation, rotation, and scaling. We observed that using data augmentation can increase the performance of the model. The proposed model was trained using different training epochs, batch sizes and dropouts. Compared with popular transfer learning approaches, the proposed model achieves better performance when using the validation data. After an extensive simulation, the proposed model achieves 96.46% classification accuracy. This accuracy of the proposed work is greater than the accuracy of traditional machine learning approaches. The proposed model is also tested with respect to its consistency and reliability. [ABSTRACT FROM AUTHOR]

Abstract In this paper, we present a deep learning-based recognition algorithm to identify pulsars by observing data containing millions of candidates including radio frequency interference and noise sources. The dataset is obtained from the High Time Resolution Universe survey created and updated by the Parkes telescope. We investigate several effective single and combined features via simple logistic regression. To deal with the imbalanced dataset, we oversimplify the original dataset at different sampling rates, which is also one of the learning parameters. After training the pre-processed dataset via a convolutional neural network, we provide a cross-validated evaluation of all candidates. Results show that the deep-learning based recognition algorithm can identify the pulsar and radio frequency interference signals with high accuracy. The precision and recall of radio frequency interference are both 100%, and those of pulsars are 91% and 94%, respectively. [ABSTRACT FROM AUTHOR]

Abstract To meet the requirement of online diagnosis of a cable fault, certain problems should be addressed. Therefore, in this paper, we propose an online cable fault diagnosis method. First, we establish a simulation model of an underground cable distribution system for collecting fault signals. Second, a deep belief network (DBN) is created by the deep learning theory for identifying a cable fault. Finally, we extract the characteristics of the fault signal and classify them into a large number of fault data automatically. A comparison of the results of the cable fault recognition with the proposed method and conventional shallow neural network shows that the DBN is of 97.8%, the conventional back propagation (BP) network is of 86.6%, ACCLN is of 94.1%, which demonstrate that the DBN-based cable fault recognition method has distinct advantages compared with a shallow neural network. [ABSTRACT FROM AUTHOR]

Human activity recognition (HAR) enables the recognition of the activities of daily living using signals from motion sensors integrated into mobile and wearable devices. One of the challenges is the uniqueness of each individual with his/her different characteristics. A general model trained without user data may perform poorly on specific users. Another challenge is running deep learning (DL) models on mobile and wearable devices due to their limited resources. In this paper, to cope with these two challenges, we use transfer learning to build personalized models and model compression for running DL algorithms. We examine the impact of different DL architectures, the number of layers to be fine-tuned, the amount of user training data, and the transfer to new datasets on the performance of HAR. We compare the performance of the transferred models with general and user-specific models in terms of F1 score, training and inference time. [Display omitted] • We use transfer learning (TL) to build personalized models for Human Activity Recognition (HAR). • We compress DL models using quantization to be deployed on mobile devices with limited resources. • We examine the impact of different architectures and layers to fine-tune on the performance of TL. • TL is compared with a general and user-specific model. • Using TL results in higher F1 scores and less running time. • Even if new labels are introduced, TL performs better, with similar positions and classes. • When the models are compressed, the inference time reduces with no compromise in accuracy. [ABSTRACT FROM AUTHOR]

• A novel joint model enhances the semantic segmentation in remote sensing images. • An improved UNet has the ability to quickly converge and avoid neuron death. • A modified SegNet optimizes the segmentation for images with imprecise labels. • Real-world experiments demonstrate the effectiveness. Semantic segmentation for high-resolution (HR) remote sensing (RS) images under the condition of a small number of training samples with imprecise labels is a challenging task. In this paper, we propose a novel method based on improved UNet and SegNet. First, a batch normalization layer is introduced into the original UNet to accelerate the convergence speed and an ELU activation function is selected instead of ReLU to avoid the neuron death. Second, to enhance the conventional SegNet, the encoder is reconstructed and a skip connection is designed to reuse the deep features, which is also beneficial for the network convergence. Third, a joint model is constructed by combining the improved UNet and SegNet and meanwhile a voting strategy is proposed to compute the final results. Experiments on the real-world HR RS images verify the effectiveness and superiority of the proposed method. [Display omitted] [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) platform allows physical devices to connect directly to the internet and upload data continuously. Insecure access makes IoT platforms vulnerable to different network intrusion attacks. As a result, the Intrusion Detection System (IDS) is a core component of a modern IoT platform. However, traditional IDS often follows rule-based detection where the rules can be changed and exposed to the attacker and becomes weak over time. An efficient IDS also needs to be dynamic and effective in real time. This paper proposes a deep learning-based algorithm to protect the network against Distributed Denial-of-Service (DDoS) attacks, insecure data flow, and similar network intrusions. A system architecture is designed for a cloud-based IoT framework to implement the proposed algorithm efficiently. The performance evaluation using standard datasets demonstrates that the proposed model provides an accuracy of up to 99.99%. [ABSTRACT FROM AUTHOR]

• AI-enabled Malware Detection System for Industrial Internet of Things is presented. • Double Density Discrete Wavelet Transform method is used for learning the features. • Hybrid CNN-LSTM model is used for the identification of the malware. • Higher classification accuracy is achieved using the proposed method. With the significant growth in Industrial Internet of Things (IIoT) technologies, various IIoT -based applications have emerged in the last decade. In recent years, various malware-based cyber-attacks have been reported on IIoT -based systems. Thus, this research work outlines the design of an efficient Artificial Intelligence (AI) -empowered zero-day malware detection system for IIoT. In this paper, a hybrid deep learning-based malware detection framework is proposed in which a Double-Density Discrete Wavelet Transform (D3WT) is used for feature extraction and a hybrid of Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model is used for identification and classification of malware. The assessment of the proposed framework is evaluated using three datasets such as IoT malware, Microsoft BIG-2015 and Malimg dataset. Experimental results show that the proposed model achieved 99.98% accuracy on the IoT malware, 96.97% accuracy on the Microsoft BIG-2015 and 99.96% accuracy with the Malimg dataset. [Display omitted] [ABSTRACT FROM AUTHOR]

Panicle segmentation is a critical step in crop phenotyping research to investigate physiological and structural traits of crops. Unmanned aerial vehicle-based image acquisition integrated with deep learning algorithms replaces labour-intensive field investigation and crop monitoring. The drone has limited resources, so it requires a small but efficient deep learning model. This paper proposes a novel Tiny Criss-Cross Network (TinyCCNET) deep learning model, based on criss-cross attention and a tiny approach. The criss-cross attention method captures contextual information from all pixels, enhancing segmentation accuracy, while the tiny approach is applied to the backbone network ResNet50. The proposed model outperforms the other existing models with 86.5% accuracy, 81.6% mIoU, and 12.3 GFLOPS. TinyCCNET achieves high accuracy, while being resource-efficient, making it an ideal choice for agricultural drone applications. [Display omitted] • Deep learning models applied to UAV requires a tiny model with higher performance. • TinyCCNET model based on cris cross attention and the tiny approach is proposed. • It is implemented on paddy rice imagery dataset, which consists of 8400 images. • It achieves 86.5% accuracy, 81.6% mIoU with 12.3 GFLOPS. [ABSTRACT FROM AUTHOR]