Search

Current mainstream deep learning methods for object detection are generally trained on high-quality datasets, which might have inferior performances under bad weather conditions. In the paper, a joint semantic deep learning algorithm is proposed to address object detection under foggy road conditions, which is constructed by embedding three attention modules and a 4-layer UNet multi-scale decoding module in the feature extraction module of the backbone network Faster RCNN. The algorithm differs from other object detection methods in that it is designed to solve low- and high-level joint tasks, including dehazing and object detection through end-to-end training. Furthermore, the location of the fog is learned by these attention modules to assist image recovery, the image quality is recovered by UNet decoding module for dehazing, and then the feature representations of the original image and the recovered image are fused and fed into the FPN (Feature Pyramid Network) module to achieve joint semantic learning. The joint semantic features are leveraged to push the subsequent network modules ability, and therefore make the proposed algorithm work better for the object detection task under foggy conditions in the real world. Moreover, this method and Faster RCNN have the same testing time due to the weight sharing in the feature extraction module. Extensive experiments confirm that the average accuracy of our algorithm outperforms the typical object detection algorithms and the state-of-the-art joint low- and high-level tasks algorithms for the object detection of seven kinds of objects on road traffics under normal weather or foggy conditions. [ABSTRACT FROM AUTHOR]

In the growing mass of social media data, how to efficiently extract the collection of interested concerns has become a research hotspot. Due to the large size and regularity of social media data, traditional indexing techniques are not applicable. Our "Learned Index", which is a part of social media intelligence solutions, uses mathematical principles to summarize the laws from the data. It predicts the location of the data by learning the mathematical properties of the data distribution to build the model. Although existing methods over single dimension and multi-dimension such as setting gaps are proposed to further optimize the performance of index, they do not consider the update-distribution of data. In this paper, we propose an update-distribution-aware learned index for social media data (TALI) to support update operations and handle the data sliding. In TALI, underlying data are learned through machine learning models, and a recursive hierarchical model is built. It also learns the update-distribution of data to adjust the size of each leaf node. Thus, it can more effectively support all kinds of operations in databases due to the decrease of the leaf nodes' sliding. In addition, TALI uses the model-based insertion method for bulkload and query, resulting in a small prediction error. Thus, exponential search is used to perform secondary lookup to improve query efficiency. Experiments were tested and compared on four realistic and synthetic social media datasets. Through extensive experiments, TALI performed better than the existing state-of-the-art learned index with less space occupancy on four realistic and synthetic social media datasets. [ABSTRACT FROM AUTHOR]

The participation of prosumers in demand-response programs is essential for the success of demand-side management in renewable-powered energy grids. Unfortunately, the engagement is still low due to concerns related to the privacy of their energy data used in the prediction processes. In this paper, we propose a blockchain-based distributed federated learning (FL) technique for energy-demand prediction that combines FL with blockchain to provide data privacy and trust features for energy prosumers. The privacy-sensitive energy data are stored locally at edge prosumer nodes without revealing it to third parties, with only the learned local model weights being shared using a blockchain network. The global federated model is not centralized but distributed and replicated over the blockchain overlay, ensuring the model immutability and provenance of parameter updates. We had proposed smart contracts to deal with the integration of local machine-learning prediction models with the blockchain, defining functions for the model parameters' scaling and reduction of blockchain overhead. The centralized, local-edge, and blockchain-integrated models are comparatively evaluated for prediction of energy demand 24 h ahead using a multi-layer perceptron model and the monitored energy data of several prosumers. The results show only a slight decrease in prediction accuracy in the case of blockchain-based distributed FL with reliable data privacy support compared with the centralized learning solution. [ABSTRACT FROM AUTHOR]

In recent years, deep learning has been successfully applied to medical image analysis and provided assistance to medical professionals. Machine learning is being used to offer diagnosis suggestions, identify regions of interest in images, or augment data to remove noise. Training models for such tasks require a large amount of labeled data. It is often difficult to procure such data due to the fact that these requires experts to manually label them, in addition to the privacy and legal concerns that limiting their collection. Due to this, creating self-supervision learning methods and domain-adaptation techniques dedicated to this domain is essential. This paper reviews concepts from the field of deep learning and how they have been applied to medical image analysis. We also review the current state of self-supervised learning methods and their applications to medical images. In doing so, we will also present the resource ecosystem of researchers in this field, such as datasets, evaluation methodologies, and benchmarks. [ABSTRACT FROM AUTHOR]

Educational data classification has become an effective tool for exploring the hidden pattern or relationship in educational data and predicting students' performance or teachers' competency. This study proposes a new method based on machine learning algorithms to predict the technology-integrated competency of pre-service mathematics teachers. In this paper, we modified the inertial subgradient extragradient algorithm for pseudomonotone equilibrium and proved the weak convergence theorem under some suitable conditions in Hilbert spaces. We then applied to solve data classification by extreme learning machine using the dataset comprised of the technology-integrated competency of 954 pre-service mathematics teachers in a university in northern Thailand, longitudinally collected for five years. The flexibility of our algorithm was shown by comparisons of the choice of different parameters. The performance was calculated and compared with the existing algorithms to be implemented for prediction. The results show that the proposed method achieved a classification accuracy of 81.06%. The predictions were implemented using ten attributes, including demographic information, skills, and knowledge relating to technology developed throughout the teacher education program. Such data driven studies are significant for establishing a prospective teacher competency analysis framework in teacher education and contributing to decision-making for policy design. [ABSTRACT FROM AUTHOR]

Rainfall-induced landslides represent a severe hazard around the world due to their sudden occurrence, as well as their widespread influence and runout distance. Considering the spatial variability of soil, stochastic analysis is often conducted to give a probability description of the runout. However, rainfall-induced landslides are complex and time-consuming for brute-force Monte Carlo analyses. Therefore, new methods are required to improve the efficiency of stochastic analysis. This paper presents a framework to investigate the influence and runout distance of rainfall-induced landslides with a two-step simulation approach. The complete process, from the initialization of instability to the post-failure flow, is simulated. The rainfall infiltration process and initialization of instability are first solved with a coupled hydro-mechanical finite element model. The post-failure flow is simulated using the coupled Eulerian–Lagrangian method, wherein the soil can flow freely in fixed Eulerian meshes. An equivalent-strength method is used to connect two steps by considering the effective stress of unsaturated soil. A rigorous method has been developed to accurately quantify the influence and runout distance via Eulerian analyses. Several simulations have been produced, using three-dimensional analyses to study the shapes of slopes and using stochastic analysis to consider uncertainty and the spatial variability of soils. It was found that a two-dimensional analysis assuming plain strain is generally conservative and safe in design, but care must be taken to interpret 2D results when the slope is convex in the longitudinal direction. The uncertainty and spatial variability of soils can lead to the statistic of influence and runout distance. The framework of using machine-learning models as surrogate models is effective in stochastic analysis of this problem and can greatly reduce computational effort. [ABSTRACT FROM AUTHOR]

The effect of the COVID-19 pandemic on crude oil prices just faded; at this moment, the Russia–Ukraine war brought a new crisis. In this paper, a new application is developed that predicts the change in crude oil prices by incorporating these two global effects. Unlike most existing studies, this work uses a dataset that involves data collected over twenty-two years and contains seven different features, such as crude oil opening, closing, intraday highest value, and intraday lowest value. This work applies cross-validation to predict the crude oil prices by using machine learning algorithms (support vector machine, linear regression, and rain forest) and deep learning algorithms (long short-term memory and bidirectional long short-term memory). The results obtained by machine learning and deep learning algorithms are compared. Lastly, the high-performance estimation can be achieved in this work with the average mean absolute error value over 0.3786. [ABSTRACT FROM AUTHOR]

Large-scale deep learning models have achieved impressive results on a variety of tasks; however, their deployment on edge or mobile devices is still a challenge due to the limited available memory and computational capability. Knowledge distillation is an effective model compression technique, which can boost the performance of a lightweight student network by transferring the knowledge from a more complex model or an ensemble of models. Due to its reduced size, this lightweight model is more suitable for deployment on edge devices. In this paper, we introduce an online knowledge distillation framework, which relies on an original attention mechanism to effectively combine the predictions of a cohort of lightweight (student) networks into a powerful ensemble, and use this as a distillation signal. The proposed aggregation strategy uses the predictions of the individual students as well as ground truth data to determine a set of weights needed for ensembling these predictions. This mechanism is solely used during system training. When testing or at inference time, a single, lightweight student is extracted and used. The extensive experiments we performed on several image classification benchmarks, both by training models from scratch (on CIFAR-10, CIFAR-100, and Tiny ImageNet datasets) and using transfer learning (on Oxford Pets and Oxford Flowers datasets), showed that the proposed framework always leads to an improvement in the accuracy of knowledge-distilled students and demonstrates the effectiveness of the proposed solution. Moreover, in the case of ResNet architecture, we observed that the knowledge-distilled model achieves a higher accuracy than a deeper, individually trained ResNet model. [ABSTRACT FROM AUTHOR]

A vehicular ad hoc network (VANET) is the major element of the intelligent transportation system (ITS). The purpose of ITS is to increase road safety and manage the movement of vehicles. ITS is known as one of the main components of smart cities. As a result, there are critical challenges such as routing in these networks. Recently, many scholars have worked on this challenge in VANET. They have used machine learning techniques to learn the routing proceeding in the networks adaptively and independently. In this paper, a Q-learning and fuzzy logic-based hierarchical routing protocol (QFHR) is proposed for VANETs. This hierarchical routing technique consists of three main phases: identifying traffic conditions, routing algorithm at the intersection level, and routing algorithm at the road level. In the first phase, each roadside unit (RSU) stores a traffic table, which includes information about the traffic conditions related to four road sections connected to the corresponding intersection. Then, RSUs use a Q-learning-based routing method to discover the best path between different intersections. Finally, vehicles in each road section use a fuzzy logic-based routing technique to choose the foremost relay node. The simulation of QFHR has been executed on the network simulator version 2 (NS2), and its results have been presented in comparison with IRQ, IV2XQ, QGrid, and GPSR in two scenarios. The first scenario analyzes the result based on the packet sending rate (PSR). In this scenario, QFHR gets better the packet delivery rate by 2.74%, 6.67%, 22.35%, and 29.98% and decreases delay by 16.19%, 22.82%, 34.15%, and 59.51%, and lowers the number of hops by 6.74%, 20.09%, 2.68%, and 12.22% compared to IRQ, IV2XQ, QGrid, and GPSR, respectively. However, it increases the overhead by approximately 9.36% and 11.34% compared to IRQ and IV2XQ, respectively. Moreover, the second scenario evaluates the results with regard to the signal transmission radius (STR). In this scenario, QFHR increases PDR by 3.45%, 8%, 23.29%, and 26.17% and decreases delay by 19.86%, 34.26%, 44.09%, and 68.39% and reduces the number of hops by 14.13%, 32.58%, 7.71%, and 21.39% compared to IRQ, IV2XQ, QGrid, and GPSR, respectively. However, it has higher overhead than IRQ (11.26%) and IV2XQ (25%). [ABSTRACT FROM AUTHOR]

One of the most challenging tasks in project management is estimating the duration of a project. The unknowns that accompany projects, the different risks, the uniqueness of each project, and the differences between organizations' culture and management techniques, hinder the ability to build one project duration prediction tool that can fit all types of projects and organizations. When machine learning (ML) techniques are used for project duration prediction, the challenge is even greater, as each organization has a different dataset structure, different features, and different quality of data. This hinders the ability to create one ML model that fits all types of organizations. This paper presents a new dynamic ML tool for improving the prediction accuracy of project duration. The tool is based on an artificial neural network (ANN) which is automatically adapted and optimized to different types of prediction methods and different datasets. The tool trains the ANN model multiple times with different architectures and uses a genetic algorithm to eventually choose the architecture which gives the most accurate prediction results. The validation process of the prediction accuracy is performed by using real-life project datasets supplied by two different organizations which have different project management approaches, different project types, and different project features. The results show that the proposed tool significantly improved the prediction accuracy for both organizations despite the major differences in the size, type, and structure of their datasets. [ABSTRACT FROM AUTHOR]

Recently, it has been demonstrated that the performance of an object detection network can be improved by embedding an attention module into it. In this work, we propose a lightweight and effective attention mechanism named multibranch attention (M3Att). For the input feature map, our M3Att first uses the grouped convolutional layer with a pyramid structure for feature extraction, and then calculates channel attention and spatial attention simultaneously and fuses them to obtain more complementary features. It is a "plug and play" module that can be easily added to the object detection network and significantly improves the performance of the object detection network with a small increase in parameters. We demonstrate the effectiveness of M3Att on various challenging object detection tasks, including PASCAL VOC2007, PASCAL VOC2012, KITTI, and Zhanjiang Underwater Robot Competition. The experimental results show that this method dramatically improves the object detection effect, especially for the PASCAL VOC2007, and the mapping index of the original network increased by 4.93% when embedded in the YOLOV4 (You Only Look Once v4) network. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs) are widely used in various artificial intelligence applications and platforms, such as sensors in internet of things (IoT) devices, speech and image recognition in mobile systems, and web searching in data centers. While DNNs achieve remarkable prediction accuracy, they introduce major computational and memory bandwidth challenges due to the increasing model complexity and the growing amount of data used for training and inference. These challenges introduce major difficulties not only due to the constraints of system cost, performance, and energy consumption, but also due to limitations in currently available memory bandwidth. The recent advances in semiconductor technologies have further intensified the gap between computational hardware performance and memory systems bandwidth. Consequently, memory systems are, today, a major performance bottleneck for DNN applications. In this paper, we present DRAMA, a deep neural network memory simulator. DRAMA extends the SCALE-Sim simulator for DNN inference on systolic arrays with a detailed, accurate, and extensive modeling and simulation environment of the memory system. DRAMA can simulate in detail the hierarchical main memory components—such as memory channels, modules, ranks, and banks—and related timing parameters. In addition, DRAMA can explore tradeoffs for memory system performance and identify bottlenecks for different DNNs and memory architectures. We demonstrate DRAMA's capabilities through a set of experimental simulations based on several use cases. [ABSTRACT FROM AUTHOR]

A control system of an autonomous robot produces a control signal based on feedback. This type of control implies the control of an object according to its state that is mathematically the control synthesis problem. Today there are no universal analytical methods for solving the general synthesis problem, and it is solved by certain particular approaches depending on the type of control object. In this paper, we propose a universal numerical approach to solving the problem of optimal control with feedback using machine learning methods based on symbolic regression. The approach is universal and can be applied to various objects. However, the use of machine learning methods imposes two aspects. First, when using them, it is necessary to reduce the requirements for optimality. In machine learning, optimization algorithms are used, but strictly optimal solutions are not sought. Secondly, in machine learning, analytical proofs of the received properties of solutions are not required. In machine methods, a set of tests is carried out and it is shown that this is sufficient to achieve the required properties. Thus, in this article, we initially introduce the fundamentals of machine learning control, introduce the basic concepts, properties and machine criteria for application of this technique. Then, with regard to the introduced notations, the feedback optimal control problem is considered and reformulated in order to add to the problem statement that such a property adjusts both the requirements of stability and optimality. Next, a description of the proposed approach is presented, theoretical formulations are given, and its efficiency is demonstrated on the computational examples in mobile robot control tasks. [ABSTRACT FROM AUTHOR]

As an important data analysis method in the field of machine learning and data mining, feature learning has a wide range of applications in various industries. The traditional multidimensional scaling (MDS) maintains the topology of data points in the low-dimensional embeddings obtained during feature learning, but ignores the discriminative nature between classes of low-dimensional embedded data. Thus, the discriminative multidimensional scaling based on pairwise constraints for feature learning (pcDMDS) model is proposed in this paper. The model enhances the discriminativeness from two aspects. The first aspect is to increase the compactness of the new data representation in the same cluster through fuzzy k-means. The second aspect is to obtain more extended pairwise constraint information between samples. In the whole feature learning process, the model considers both the topology of samples in the original space and the cluster structure in the new space. It also incorporates the extended pairwise constraint information in the samples, which further improves the model's ability to obtain discriminative features. Finally, the experimental results on twelve datasets show that pcDMDS performs 10.31 % and 8.31 % higher than PMDS model in terms of accuracy and purity. [ABSTRACT FROM AUTHOR]

Companies, institutions or governments process large amounts of data for the development of their activities. This knowledge usually comes from devices that collect data from various sources. Processing them in real time is essential to ensure the flow of information about the current state of infrastructure, as this knowledge is the basis for management and decision making in the event of an attack or anomalous situations. Therefore, this article exposes three unsupervised machine learning models based on clustering techniques and threshold definitions to detect anomalies from heterogeneous streaming cybersecurity data sources. After evaluation, this paper presents a case of heterogeneous cybersecurity devices, comparing WSSSE, Silhouette and training time metrics for all models, where K-Means was defined as the optimal algorithm for anomaly detection in streaming data processing. The anomaly detection's accuracy achieved is also significantly high. A comparison with other research studies is also performed, against which the proposed method proved its strong points. [ABSTRACT FROM AUTHOR]

Competing risk ( C o R ) models are frequently disregarded in failure rate analysis, and traditional statistical approaches are used to study the event of interest. In this paper, we proposed a new lifetime distribution by generalizing the length biased exponential (LBE) distribution using the transmuted Topp-Leone-G ( T T L -G) family of distributions. The new three parameter model is called the transmuted Topp-Leone length biased exponential ( T T L L B E ) distribution. A comprehensive account of various mathematical features of the T T L L B E model are derived. The unknown parameters of the proposed distribution are estimated by six classical approaches: the maximum likelihood (ML) approach, maximum product spacing (MPS) approach, least square (LS) approach, Weighted LS (WLS) approach, Cramér-Von Mises (CVN) approach, Anderson–Darling (AD) approach, and Bayesian approach. The stability of the model parameters is examined through the simulation study. The applications of our proposed distribution are explained through real data and its performance is illustrated through its comparison with the competent existing distributions. The T T L L B E model depend on the C o R model has been obtained and estimated parameter of this model by ML and Bayesian estimation approaches. In electrical appliances, we found two main causes of failure, and the data of electrical appliances are fitted to our model. Therefore, we analyzed the T T L L B E model depend on the C o R model to obtain the strong cause of failure. [ABSTRACT FROM AUTHOR]

Recent technologies such as artificial intelligence, machine learning, and big data are essential for supporting healthcare monitoring systems, particularly for monitoring Monkeypox confirmed cases. Infected and uninfected cases around the world have contributed to a growing dataset, which is publicly available and can be used by artificial intelligence and machine learning to predict the confirmed cases of Monkeypox at an early stage. Motivated by this, we propose in this paper a new approach for accurate prediction of the Monkeypox confirmed cases based on an optimized Long Short-Term Memory (LSTM) deep network. To fine-tune the hyper-parameters of the LSTM-based deep network, we employed the Al-Biruni Earth Radius (BER) optimization algorithm; thus, the proposed approach is denoted by BER-LSTM. Experimental results show the effectiveness of the proposed approach when assessed using various evaluation criteria, such as Mean Bias Error, which is recorded as (0.06) using BER-LSTM. To prove the superiority of the proposed approach, six different machine learning models are included in the conducted experiments. In addition, four different optimization algorithms are considered for comparison purposes. The results of this comparison confirmed the superiority of the proposed approach. On the other hand, several statistical tests are applied to analyze the stability and significance of the proposed approach. These tests include one-way Analysis of Variance (ANOVA), Wilcoxon, and regression tests. The results of these tests emphasize the robustness, significance, and efficiency of the proposed approach. [ABSTRACT FROM AUTHOR]

Climate disasters such as floods and droughts often bring heavy losses to human life, national economy, and public safety. El Niño/Southern Oscillation (ENSO) is one of the most important inter-annual climate signals in the tropics and has a global impact on atmospheric circulation and precipitation. To address the impact of climate change, accurate ENSO forecasts can help prevent related climate disasters. Traditional prediction methods mainly include statistical methods and dynamic methods. However, due to the variability and diversity of the temporal and spatial evolution of ENSO, traditional methods still have great uncertainty in predicting ENSO. In recent years, with the rapid development of artificial intelligence technology, it has gradually penetrated into all aspects of people's lives, and the climate field has also benefited. For example, deep learning methods in artificial intelligence can automatically learn and train from a large amount of sample data, obtain excellent feature representation, and effectively improve the performance of various learning tasks. It is widely used in computer vision, natural language processing, and other fields. In 2019, Ham et al. used a convolutional neural network (CNN) model in ENSO forecasting 18 months in advance, and the winter ENSO forecasting skill could reach 0.64, far exceeding the dynamic model with a forecasting skill of 0.5. The research results were regarded as the pioneering work of deep learning in the field of weather forecasting. This paper introduces the traditional ENSO forecasting methods and focuses on summarizing the various latest artificial intelligence methods and their forecasting effects for ENSO forecasting, so as to provide useful reference for future research by researchers. [ABSTRACT FROM AUTHOR]

This paper aims at performing a comparative study to investigate the predictive capability of machine learning (ML) models used for estimating the compressive strength of self-compacting concrete (SCC). Seven prominent ML models, including deep neural network regression (DNNR), extreme gradient boosting machine (XGBoost), gradient boosting machine (GBM), adaptive boosting machine (AdaBoost), support vector regression (SVR), Levenberg–Marquardt artificial neural network (LM-ANN), and genetic programming (GP), are employed. Four experimental datasets, compiled in previous studies, are used to construct the ML-based methods. The models' generalization capabilities are reliably evaluated by 20 independent runs. Experimental results point out the superiority of the DNNR, which has excelled other models in three out of four datasets. The XGBoost is the second-best model, which has gained the first rank in one dataset. The outcomes point out the great potential of the utilized ML approaches in modeling the compressive strength of SCC. In more details, the coefficient of determination (R2) surpasses 0.8 and the mean absolute percentage error (MAPE) is always below 15% for all datasets. The best results of R2 and MAPE are 0.93 and 7.2%, respectively. [ABSTRACT FROM AUTHOR]

This paper proposes a clustering method based on a randomized representation of an ensemble of possible clusters with a probability distribution. The concept of a cluster indicator is introduced as the average distance between the objects included in the cluster. The indicators averaged over the entire ensemble are considered the latter's characteristics. The optimal distribution of clusters is determined using the randomized machine learning approach: an entropy functional is maximized with respect to the probability distribution subject to constraints imposed on the averaged indicator of the cluster ensemble. The resulting entropy-optimal cluster corresponds to the maximum of the optimal probability distribution. This method is developed for binary clustering as a basic procedure. Its extension to t-ary clustering is considered. Some illustrative examples of entropy-randomized clustering are given. [ABSTRACT FROM AUTHOR]

An organization's success depends on its employees, and an employee's performance decides whether the organization is successful. Employee performance enhances the productivity and output of organizations, i.e., the performance of an employee paves the way for the organization's success. Hence, analyzing employee performance and giving performance ratings to employees is essential for companies nowadays. It is evident that different people have different skill sets and behavior, so data should be gathered from all parts of an employee's life. This paper aims to provide the performance rating of an employee based on various factors. First, we compare various AI-based algorithms, such as random forest, artificial neural network, decision tree, and XGBoost. Then, we propose an ensemble approach, RanKer, combining all the above approaches. The empirical results illustrate that the efficacy of the proposed model compared to traditional models such as random forest, artificial neural network, decision tree, and XGBoost is high in terms of precision, recall, F1-score, and accuracy. [ABSTRACT FROM AUTHOR]

Controlling time-delayed processes is one of the challenges in today's process industries. If the multi-input/multi-output system is dynamically coupled, the delay problem becomes more critical. In this paper, a new method based on Smith's predictive method, with the help of a type-2 fuzzy system to control the system with the mentioned features, is presented. The variability in the time delay, the existence of disturbances and the existence of structural and parametric uncertainty lead to the poor performance of the traditional Smith predictor. Even if the control system is set up correctly at the beginning of the setup, it will eventually wear out, and the above problems will appear. Therefore, computational intelligence is used here, and by updating the parameters of the control system at the same time as the system changes, the control system adapts itself to achieve the best performance. To evaluate the proposed control system, a complex process system is simulated, the results of which show the good performance of Smith's prediction method based on a type-2 fuzzy system. [ABSTRACT FROM AUTHOR]

With the booming development of bridge construction, bridge operation and maintenance have always been major issues to ensure the safety of the community. Affected by the long-term service of bridges and natural factors, the safety and durability of cables can be threatened. Cables are critical stress-bearing elements of large bridges such as cable-stayed bridges. Realizing the health monitoring of bridge cables is the key to ensuring the normal operation of bridges. Acoustic emission (AE) is a dynamic nondestructive testing method that is increasingly used in the local monitoring of bridge cables. In this paper, a testbed is described for generating the acoustic emission signals for signal identification testing with machine learning (ML) models. Owing to the limited number of measured signals being available, an algorithm is proposed to simulate acoustic emission signals for model training. A multi-angle feature extraction method is proposed to extract the acoustic emission signals and construct a comprehensive feature vector to characterize the acoustic emission signals. Seven ML models are trained with the simulated acoustic emission signals. Long short-term memory (LSTM) has been specially applied for deep learning demonstration which requires a large amount of training data. As all machine learning models (including LSTM) provide desired performance, it shows that the proposed approach of simulating acoustic emission signals can be effective. [ABSTRACT FROM AUTHOR]

Automatic vessel structure segmentation is essential for an automatic disease diagnosis system. The task is challenging due to vessels' different shapes and sizes across populations. This paper proposes a multiscale network with dual attention to segment various retinal blood vessels. The network injects a spatial attention module and channel attention module on a feature map, whose size is one-eighth of the input size. The network also uses multiscale input to receive multi-level information, and the network uses the multiscale output to gain more supervision. The proposed method is tested on two publicly available datasets: DRIVE and CHASEDB1. The accuracy, AUC, sensitivity, and specificity on the DRIVE dataset are 0.9615, 0.9866, 0.7709, and 0.9847, respectively. On the CHASEDB1 dataset, the metrics are 0.9800, 0.9892, 0.8215, and 0.9877, respectively. The ablative study further shows effectiveness for each part of the network. Multiscale and dual attention mechanism both improve performance. The proposed architecture is simple and effective. The inference time is 12 ms on a GPU and has potential for real-world applications. The code will be made publicly available. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) offer significant advantages when used in various image classification tasks and computer vision applications. CNNs are increasingly deployed in environments from edge and Internet of Things (IoT) devices to high-end computational infrastructures, such as supercomputers, cloud computing, and data centers. The growing amount of data and the growth in their model size and computational complexity, however, introduce major computational challenges. Such challenges present entry barriers for IoT and edge devices as well as increase the operational expenses of large-scale computing systems. Thus, it has become essential to optimize CNN algorithms. In this paper, we introduce the S-VELCRO compression algorithm, which exploits value locality to trim filters in CNN models utilized for specialized tasks. S-VELCRO uses structured compression, which can save costs and reduce overhead compared with unstructured compression. The algorithm runs in two steps: a preprocessing step identifies the filters with a high degree of value locality, and a compression step trims the selected filters. As a result, S-VELCRO reduces the computational load of the channel activation function and avoids the convolution computation of the corresponding trimmed filters. Compared with typical CNN compression algorithms that run heavy back-propagation training computations, S-VELCRO has significantly fewer computational requirements. Our experimental analysis shows that S-VELCRO achieves a compression-saving ratio between 6% and 30%, with no degradation in accuracy for ResNet-18, MobileNet-V2, and GoogLeNet when used for specialized tasks. [ABSTRACT FROM AUTHOR]

The world is still trying to recover from the devastation caused by the wide spread of COVID-19, and now the monkeypox virus threatens becoming a worldwide pandemic. Although the monkeypox virus is not as lethal or infectious as COVID-19, numerous countries report new cases daily. Thus, it is not surprising that necessary precautions have not been taken, and it will not be surprising if another worldwide pandemic occurs. Machine learning has recently shown tremendous promise in image-based diagnosis, including cancer detection, tumor cell identification, and COVID-19 patient detection. Therefore, a similar application may be implemented to diagnose monkeypox as it invades the human skin. An image can be acquired and utilized to further diagnose the condition. In this paper, two algorithms are proposed for improving the classification accuracy of monkeypox images. The proposed algorithms are based on transfer learning for feature extraction and meta-heuristic optimization for feature selection and optimization of the parameters of a multi-layer neural network. The GoogleNet deep network is adopted for feature extraction, and the utilized meta-heuristic optimization algorithms are the Al-Biruni Earth radius algorithm, the sine cosine algorithm, and the particle swarm optimization algorithm. Based on these algorithms, a new binary hybrid algorithm is proposed for feature selection, along with a new hybrid algorithm for optimizing the parameters of the neural network. To evaluate the proposed algorithms, a publicly available dataset is employed. The assessment of the proposed optimization of feature selection for monkeypox classification was performed in terms of ten evaluation criteria. In addition, a set of statistical tests was conducted to measure the effectiveness, significance, and robustness of the proposed algorithms. The results achieved confirm the superiority and effectiveness of the proposed methods compared to other optimization methods. The average classification accuracy was 98.8%. [ABSTRACT FROM AUTHOR]

A common assumption in machine learning is that training data is complete, and the data distribution is fixed. However, in many practical applications, this assumption does not hold. Incremental learning was proposed to compensate for this problem. Common approaches include retraining models and incremental learning to compensate for the shortage of training data. Retraining models is time-consuming and computationally expensive, while incremental learning can save time and computational costs. However, the concept drift may affect the performance. Two crucial issues should be considered to address concept drift in incremental learning: gaining new knowledge without forgetting previously acquired knowledge and forgetting obsolete information without corrupting valid information. This paper proposes an incremental support vector machine learning approach with domain adaptation, considering both crucial issues. Firstly, a small amount of new data is used to fine-tune the previous model to generate a model that is sensitive to the new data but retains the previous data information by transferring parameters. Secondly, an ensemble and model selection mechanism based on Bayesian theory is proposed to keep the valid information. The computational experiments indicate that the performance of the proposed model improved as new data was acquired. In addition, the influence of the degree of data drift on the algorithm is also explored. A gain in performance on four out of five industrial datasets and four synthetic datasets has been demonstrated over the support vector machine and incremental support vector machine algorithms. [ABSTRACT FROM AUTHOR]

The fine particulate matter (PM2.5) concentration has been a vital source of info and an essential indicator for measuring and studying the concentration of other air pollutants. It is crucial to realize more accurate predictions of PM2.5 and establish a high-accuracy PM2.5 prediction model due to their social impacts and cross-field applications in geospatial engineering. To further boost the accuracy of PM2.5 prediction results, this paper proposes a new wavelet PM2.5 prediction system (called WD-OSMSSA-KELM model) based on a new, improved variant of the salp swarm algorithm (OSMSSA), kernel extreme learning machine (KELM), wavelet decomposition, and Boruta-XGBoost (B-XGB) feature selection. First, we applied the B-XGB feature selection to realize the best features for predicting hourly PM2.5 concentrations. Then, we applied the wavelet decomposition (WD) algorithm to reach the multi-scale decomposition results and single-branch reconstruction of PM2.5 concentrations to mitigate the prediction error produced by time series data. In the next stage, we optimized the parameters of the KELM model under each reconstructed component. An improved version of the SSA is proposed to reach higher performance for the basic SSA optimizer and avoid local stagnation problems. In this work, we propose new operators based on oppositional-based learning and simplex-based search to mitigate the core problems of the conventional SSA. In addition, we utilized a time-varying parameter instead of the main parameter of the SSA. To further boost the exploration trends of SSA, we propose using the random leaders to guide the swarm towards new regions of the feature space based on a conditional structure. After optimizing the model, the optimized model was utilized to predict the PM2.5 concentrations, and different error metrics were applied to evaluate the model's performance and accuracy. The proposed model was evaluated based on an hourly database, six air pollutants, and six meteorological features collected from the Beijing Municipal Environmental Monitoring Center. The experimental results show that the proposed WD-OLMSSA-KELM model can predict the PM2.5 concentration with superior performance (R: 0.995, RMSE: 11.906, MdAE: 2.424, MAPE: 9.768, KGE: 0.963, R 2 : 0.990) compared to the WD-CatBoost, WD-LightGBM, WD-Xgboost, and WD-Ridge methods. [ABSTRACT FROM AUTHOR]

Uniaxial compressive strength (UCS) is a critical parameter in the disaster prevention of engineering projects, requiring a large budget and a long time to estimate in different rocks or the early stage of a project. If predicted accurately, the UCS of rocks significantly affects geotechnical applications. This paper develops a dataset of 734 samples from previous studies on different countries' magmatic, sedimentary, and metamorphic rocks. Within the study context, three main factors, point load index, P-wave velocity, and Schmidt hammer rebound number, are utilized to estimate UCS. Moreover, it applies extreme learning machines (ELM) to map the nonlinear relationship between the UCS and the influential factors. Five metaheuristic algorithms, particle swarm optimization (PSO), grey wolf optimization (GWO), whale optimization algorithm (WOA), butterfly optimization algorithm (BOA), and sparrow search algorithm (SSA), are used to optimize the bias and weight of ELM and thus enhance its predictability. Indeed, several performance parameters are utilized to verify the proposed models' generalization capability and predictive performance. The minimum, maximum, and average relative errors of ELM achieved by the whale optimization algorithm (WOA-ELM) are smaller than the other models, with values of 0.22%, 72.05%, and 11.48%, respectively. In contrast, the minimum and mean residual error produced by WOA-ELM are less than the other models, with values of 0.02 and 2.64 MPa, respectively. The results show that the UCS values derived from WOA-ELM are superior to those from other models. The performance indices (coefficient of determination (R2): 0.861, mean squared error (MSE): 17.61, root mean squared error (RMSE): 4.20, and value account for (VAF): 91% obtained using the WOA-ELM model indicates high accuracy and reliability, which means that it has broad application potential for estimating UCS of different rocks. [ABSTRACT FROM AUTHOR]

In recent years, with the rapid development of Internet services in all walks of life, a large number of malicious acts such as network attacks, data leakage, and information theft have become major challenges for network security. Due to the difficulty of malicious traffic collection and labeling, the distribution of various samples in the existing dataset is seriously imbalanced, resulting in low accuracy of malicious traffic classification based on machine learning and deep learning, and poor model generalization ability. In this paper, a feature image representation method and Adversarial Generative Network with Filter (Filter-GAN) are proposed to solve these problems. First, the feature image representation method divides the original session traffic into three parts. The Markov matrix is extracted from each part to form a three-channel feature image. This method can transform the original session traffic format into a uniform-length matrix and fully characterize the network traffic. Then, Filter-GAN uses the feature images to generate few attack samples. Compared with general methods, Filter-GAN can generate more efficient samples. Experiments were conducted on public datasets. The results show that the feature image representation method can effectively characterize the original session traffic. When the number of samples is sufficient, the classification accuracy can reach 99%. Compared with unbalanced datasets, Filter-GAN has significantly improved the recognition accuracy of small-sample datasets, with a maximum improvement of 6%. [ABSTRACT FROM AUTHOR]

In line with current developments, biometrics is becoming an important technology that enables safer identification of individuals and more secure access to sensitive information and assets. Researchers have recently started exploring electroencephalography (EEG) as a biometric modality thanks to the uniqueness of EEG signals. A new architecture for a convolutional neural network (CNN) that uses EEG signals is suggested in this paper for biometric identification. A CNN does not need complex signal pre-processing, feature extraction, and feature selection stages. The EEG datasets utilized in this research are the resting state eyes open (REO) and the resting state eyes closed (REC) EEG. Extensive experiments were performed to design this deep CNN architecture. These experiments showed that a CNN architecture with eleven layers (eight convolutional layers, one average pooling layer, and two fully connected layers) with an Adam optimizer resulted in the highest accuracy. The CNN architecture proposed here was compared to existing models for biometrics using the same dataset. The results show that the proposed method outperforms the other task-free paradigm CNN biometric identification models, with an identification accuracy of 98.54%. [ABSTRACT FROM AUTHOR]

In this paper, we study the joint optimization problem of the spectrum and power allocation for multiple vehicle-to-infrastructure (V2I) and vehicle-to-vehicle (V2V) users in cellular vehicle-to-everything (C-V2X) communication, aiming to maximize the sum rate of V2I links while satisfying the low latency requirements of V2V links. However, channel state information (CSI) is hard to obtain accurately due to the mobility of vehicles. In addition, the effective sensing of state information among vehicles becomes difficult in an environment with complex and diverse information, which is detrimental to vehicles collaborating for resource allocation. Thus, we propose a framework of multi-agent deep reinforcement learning based on attention mechanism (AMARL) to improve the V2X communication performance. Specifically, for vehicle mobility, we model the problem as a multi-agent reinforcement learning process, where each V2V link is regarded an agent and all agents jointly intercommunicate with the environment. Each agent allocates spectrum and power through its deep Q network (DQN). To enhance effective intercommunication and the sense of collaboration among vehicles, we introduce an attention mechanism to focus on more relevant information, which in turn reduces the signaling overhead and optimizes their communication performance more explicitly. Experimental results show that the proposed AMARL-based approach can satisfy the requirements of a high rate for V2I links and low latency for V2V links. It also has an excellent adaptability to environmental change. [ABSTRACT FROM AUTHOR]

Mental fatigue is a major public health issue worldwide that is common among both healthy and sick people. In the literature, various modern technologies, together with artificial intelligence techniques, have been proposed. Most techniques consider complex biosignals, such as electroencephalogram, electro-oculogram or classification of basic heart rate variability parameters. Additionally, most studies focus on a particular area, such as driving, surgery, etc. In this paper, a novel approach is presented that combines electrocardiogram (ECG) signal feature extraction, principal component analysis (PCA), and classification using machine learning algorithms. With the aim of daily mental fatigue recognition, an experiment was designed wherein ECG signals were recorded twice a day: in the morning, i.e., a state without fatigue, and in the evening, i.e., a fatigued state. PCA analysis results show that ECG signal parameters, such as Q and R wave amplitude values, as well as QT and T intervals, presented with the largest differences between states compared to other ECG signal parameters. Furthermore, the random forest classifier achieved more than 94.5% accuracy. This work demonstrates the feasibility of ECG signal feature extraction for automatic mental fatigue detection. [ABSTRACT FROM AUTHOR]

University dropout rates are a problem that presents many negative consequences. It is an academic issue and carries an unfavorable economic impact. In recent years, significant efforts have been devoted to the early detection of students likely to drop out. This paper uses data corresponding to dropout candidates after their first year in the third largest face-to-face university in Europe, with the goal of predicting likely dropout either at the beginning of the course of study or at the end of the first semester. In this prediction, we considered the five major program areas. Different techniques have been used: first, a Feature Selection Process in order to identify the variables more correlated with dropout; then, some Machine Learning Models (Support Vector Machines, Decision Trees and Artificial Neural Networks) as well as a Logistic Regression. The results show that dropout detection does not work only with enrollment variables, but it improves after the first semester results. Academic performance is always a relevant variable, but there are others, such as the level of preference that the student had over the course that he or she was finally able to study. The success of the techniques depends on the program areas. Machine Learning obtains the best results, but a simple Logistic Regression model can be used as a reasonable baseline. [ABSTRACT FROM AUTHOR]

When training a feedforward stochastic gradient descendent trained neural network, there is a possibility of not learning a batch of patterns correctly that causes the network to fail in the predictions in the areas adjacent to those patterns. This problem has usually been resolved by directly adding more complexity to the network, normally by increasing the number of learning layers, which means it will be heavier to run on the workstation. In this paper, the properties and the effect of the patterns on the network are analysed and two main reasons why the patterns are not learned correctly are distinguished: the disappearance of the Jacobian gradient on the processing layers of the network and the opposite direction of the gradient of those patterns. A simplified experiment has been carried out on a simple neural network and the errors appearing during and after training have been monitored. Taking into account the data obtained, the initial hypothesis of causes seems to be correct. Finally, some corrections to the network are proposed with the aim of solving those training issues and to be able to offer a sufficiently correct prediction, in order to increase the complexity of the network as little as possible. [ABSTRACT FROM AUTHOR]

Wheeled mobile robots are widely used for their excellent mobility and high maneuverability. The intelligentization of mobile robots to perform autonomous movement is one of the favorite topics in the robotic field. This paper proposes a new intelligent path-following scheme for mobile robots. A secure path is designed by the chaotic systems and an unknown switching mechanism. The proposed controller is constructed by the type-3 (T3) fuzzy logic systems (FLSs) and a predictive compensator. The T3-FLSs are taught to obtain good accuracy. In addition, the stability is mathematically investigated and guaranteed by the designed compensator. The simulations show that the case-study robot follows the planned secure path well and resists uncertainties. [ABSTRACT FROM AUTHOR]

This paper proposes a novel hybrid optimization technique based on a machine learning (ML) approach and transient search optimization (TSO) to solve the optimal power flow problem. First, the study aims at developing and evaluating the proposed hybrid ML-TSO algorithm. To do so, the optimization technique is implemented to solve the classical optimal power flow problem (OPF), with an objective function formulated to minimize the total generation costs. Second, the hybrid ML-TSO is adapted to solve the probabilistic OPF problem by studying the impact of the unavoidable uncertainty of renewable energy sources (solar photovoltaic and wind turbines) and time-varying load profiles on the generation costs. The evaluation of the proposed solution method is examined and validated on IEEE 57-bus and 118-bus standard systems. The simulation results and comparisons confirmed the robustness and applicability of the proposed hybrid ML-TSO algorithm in solving the classical and probabilistic OPF problems. Meanwhile, a significant reduction in the generation costs is attained upon the integration of the solar and wind sources into the investigated power systems. [ABSTRACT FROM AUTHOR]

In recent years, flying ad hoc networks have attracted the attention of many researchers in industry and universities due to easy deployment, proper operational costs, and diverse applications. Designing an efficient routing protocol is challenging due to unique characteristics of these networks such as very fast motion of nodes, frequent changes of topology, and low density. Routing protocols determine how to provide communications between drones in a wireless ad hoc network. Today, reinforcement learning (RL) provides powerful solutions to solve the existing problems in the routing protocols, and designs autonomous, adaptive, and self-learning routing protocols. The main purpose of these routing protocols is to ensure a stable routing solution with low delay and minimum energy consumption. In this paper, the reinforcement learning-based routing methods in FANET are surveyed and studied. Initially, reinforcement learning, the Markov decision process (MDP), and reinforcement learning algorithms are briefly described. Then, flying ad hoc networks, various types of drones, and their applications, are introduced. Furthermore, the routing process and its challenges are briefly explained in FANET. Then, a classification of reinforcement learning-based routing protocols is suggested for the flying ad hoc networks. This classification categorizes routing protocols based on the learning algorithm, the routing algorithm, and the data dissemination process. Finally, we present the existing opportunities and challenges in this field to provide a detailed and accurate view for researchers to be aware of the future research directions in order to improve the existing reinforcement learning-based routing algorithms. [ABSTRACT FROM AUTHOR]

Alternating Direction Method of Multipliers (ADMM) is a widely used machine learning tool in distributed environments. In the paper, we propose an ADMM-based differential privacy learning algorithm (FDP-ADMM) on penalized quantile regression for distributed functional data. The FDP-ADMM algorithm can resist adversary attacks to avoid the possible privacy leakage in distributed networks, which is designed by functional principal analysis, an approximate augmented Lagrange function, ADMM algorithm, and privacy policy via Gaussian mechanism with time-varying variance. It is also a noise-resilient, convergent, and computationally effective distributed learning algorithm, even if for high privacy protection. The theoretical analysis on privacy and convergence guarantees is derived and offers a privacy–utility trade-off: a weaker privacy guarantee would result in better utility. The evaluations on simulation-distributed functional datasets have demonstrated the effectiveness of the FDP-ADMM algorithm even if under high privacy guarantee. [ABSTRACT FROM AUTHOR]

Policymakers and researchers worldwide are interested in measuring the subjective well-being (SWB) of populations. In recent years, new approaches to measuring SWB have begun to appear, using digital traces as the main source of information, and show potential to overcome the shortcomings of traditional survey-based methods. In this paper, we propose the formal model for calculation of observable subjective well-being (OSWB) indicator based on posts from a social network, which utilizes demographic information and post-stratification techniques to make the data sample representative by selected characteristics of the general population. We applied the model on the data from Odnoklassniki, one of the largest social networks in Russia, and obtained an OSWB indicator representative of the population of Russia by age and gender. For sentiment analysis, we fine-tuned several language models on RuSentiment and achieved state-of-the-art results. The calculated OSWB indicator demonstrated moderate to strong Pearson's ( r = 0.733 , p = 0.007 , n = 12 ) correlation and strong Spearman's ( r s = 0.825 , p = 0.001 , n = 12 ) correlation with a traditional survey-based Happiness Index reported by Russia Public Opinion Research Center, confirming the validity of the proposed approach. Additionally, we explored circadian (24 h) and circaseptan (7 day) patterns, and report several interesting findings for the population of Russia. Firstly, daily variations were clearly observed: the morning had the lowest level of happiness, and the late evening had the highest. Secondly, weekly patterns were clearly observed as well, with weekends being happier than weekdays. The lowest level of happiness occurs in the first three weekdays, and starting on Thursday, it rises and peaks during the weekend. Lastly, demographic groups showed different levels of happiness on a daily, weekly, and monthly basis, which confirms the importance of post-stratification by age group and gender in OSWB studies based on digital traces. [ABSTRACT FROM AUTHOR]

Evolutionary approaches have a critical role in different disciplines such as real-world problems, computer programming, machine learning, biological sciences, and many more. The design of the stochastic model is based on transition probabilities and non-parametric techniques. Positivity, boundedness, and equilibria are investigated in deterministic and stochastic senses. An essential tool, Euler–Maruyama, is studied for the solution of said model. Standard and nonstandard evolutionary approaches are presented for the stochastic model in terms of efficiency and low-cost approximations. The standard evolutionary procedures like stochastic Euler–Maruyama and stochastic Runge–Kutta fail to restore the essential features of biological problems. On the other hand, the proposed method is efficient, of meager cost, and adopts all the desired feasible properties. At the end of this paper the comparison section is presented to support efficient analysis. [ABSTRACT FROM AUTHOR]

In this paper, we propose an extrinsic approach based on physics-informed neural networks (PINNs) for solving the partial differential equations (PDEs) on surfaces embedded in high dimensional space. PINNs are one of the deep learning-based techniques. Based on the training data and physical models, PINNs introduce the standard feedforward neural networks (NNs) to approximate the solutions to the PDE systems. Using automatic differentiation, the PDEs information could be encoded into NNs and a loss function. To deal with the surface differential operators in the loss function, we combine the extrinsic approach with PINNs and then express that loss function in extrinsic form. Subsequently, the loss function could be minimized extrinsically with respect to the NN parameters. Numerical results demonstrate that the extrinsic approach based on PINNs for surface problems has good accuracy and higher efficiency compared with the embedding approach based on PINNs. In addition, the strong nonlinear mapping ability of NNs makes this approach robust in solving time-dependent nonlinear problems on more complex surfaces. [ABSTRACT FROM AUTHOR]

The complexity of the data type and distribution leads to the increase in uncertainty in the relationship between samples, which brings challenges to effectively mining the potential cluster structure of data. Ensemble clustering aims to obtain a unified cluster division by fusing multiple different base clustering results. This paper proposes a three-way ensemble clustering algorithm based on sample's perturbation theory to solve the problem of inaccurate decision making caused by inaccurate information or insufficient data. The algorithm first combines the natural nearest neighbor algorithm to generate two sets of perturbed data sets, randomly extracts the feature subsets of the samples, and uses the traditional clustering algorithm to obtain different base clusters. The sample's stability is obtained by using the co-association matrix and determinacy function, and then the samples can be divided into a stable region and unstable region according to a threshold for the sample's stability. The stable region consists of high-stability samples and is divided into the core region of each cluster using the K-means algorithm. The unstable region consists of low-stability samples and is assigned to the fringe regions of each cluster. Therefore, a three-way clustering result is formed. The experimental results show that the proposed algorithm in this paper can obtain better clustering results compared with other clustering ensemble algorithms on the UCI Machine Learning Repository data set, and can effectively reveal the clustering structure. [ABSTRACT FROM AUTHOR]

Many metaheuristic approaches have been developed to select effective features from different medical datasets in a feasible time. However, most of them cannot scale well to large medical datasets, where they fail to maximize the classification accuracy and simultaneously minimize the number of selected features. Therefore, this paper is devoted to developing an efficient binary version of the quantum-based avian navigation optimizer algorithm (QANA) named BQANA, utilizing the scalability of the QANA to effectively select the optimal feature subset from high-dimensional medical datasets using two different approaches. In the first approach, several binary versions of the QANA are developed using S-shaped, V-shaped, U-shaped, Z-shaped, and quadratic transfer functions to map the continuous solutions of the canonical QANA to binary ones. In the second approach, the QANA is mapped to binary space by converting each variable to 0 or 1 using a threshold. To evaluate the proposed algorithm, first, all binary versions of the QANA are assessed on different medical datasets with varied feature sizes, including Pima, HeartEW, Lymphography, SPECT Heart, PenglungEW, Parkinson, Colon, SRBCT, Leukemia, and Prostate tumor. The results show that the BQANA developed by the second approach is superior to other binary versions of the QANA to find the optimal feature subset from the medical datasets. Then, the BQANA was compared with nine well-known binary metaheuristic algorithms, and the results were statistically assessed using the Friedman test. The experimental and statistical results demonstrate that the proposed BQANA has merit for feature selection from medical datasets. [ABSTRACT FROM AUTHOR]

It is significant for anesthesiologists to have a precise grasp of the recovery time of the patient after anesthesia. Accurate prediction of anesthesia recovery time can support anesthesiologist decision-making during surgery to help reduce the risk of surgery in patients. However, effective models are not proposed to solve this problem for anesthesiologists. In this paper, we seek to find effective forecasting methods. First, we collect 1824 patient anesthesia data from the eye center and then performed data preprocessing. We extracted 85 variables to predict recovery time from anesthesia. Second, we extract anesthesia information between variables for prediction using machine learning methods, including Bayesian ridge, lightGBM, random forest, support vector regression, and extreme gradient boosting. We also design simple deep learning models as prediction models, including linear residual neural networks and jumping knowledge linear neural networks. Lastly, we perform a comparative experiment of the above methods on the dataset. The experiment demonstrates that the machine learning method performs better than the deep learning model mentioned above on a small number of samples. We find random forest and XGBoost are more efficient than other methods to extract information between variables on postoperative anesthesia recovery time. [ABSTRACT FROM AUTHOR]

In this paper, we consider the poroelasticity problem in heterogeneous media. The mathematical model is described by a coupled system of equations for displacement and pressure in the coupled dual continuum porous media. We propose a new method based on hybrid explicit–implicit (HEI) learning to solve the poroelasticity problem in dual continuum heterogeneous media. We use a finite element method with standard linear basis functions for spatial approximation. We apply the explicit–implicit time scheme, where the explicit scheme is used for the low-conductive continuum and the implicit scheme for the high-conductive. The fixed-strain splitting scheme is used to accelerate the computation and decouple the flow and mechanics problems. The main idea of the proposed method is partial learning of particular degrees of freedom of the high-conductive continuum's pressure (implicit part of the flow). First, we train a deep neural network (DNN) to obtain values of the implicit part of the flow at some spatial points at some time moments. Then, we apply the Discrete Empirical Interpolation Method (DEIM) combined with Proper Orthogonal Decomposition (POD) to restore the complete implicit parts and perform linear interpolation over time. Consequently, we treat the high-conductive continuum's pressure as a known function and use it to find the other continuum's pressure and displacements. Numerical results for the two-dimensional model problem are presented. The results demonstrate that the proposed method provides fast and accurate predictions. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) is an evolving set of technologies used for solving a wide range of applied issues. The core of AI is machine learning (ML)—a complex of algorithms and methods that address the problems of classification, clustering, and forecasting. The practical application of AI&ML holds promising prospects. Therefore, the researches in this area are intensive. However, the industrial applications of AI and its more intensive use in society are not widespread at the present time. The challenges of widespread AI applications need to be considered from both the AI (internal problems) and the societal (external problems) perspective. This consideration will identify the priority steps for more intensive practical application of AI technologies, their introduction, and involvement in industry and society. The article presents the identification and discussion of the challenges of the employment of AI technologies in the economy and society of resource-based countries. The systematization of AI&ML technologies is implemented based on publications in these areas. This systematization allows for the specification of the organizational, personnel, social and technological limitations. This paper outlines the directions of studies in AI and ML, which will allow us to overcome some of the limitations and achieve expansion of the scope of AI&ML applications. [ABSTRACT FROM AUTHOR]

This paper compares two statistical methods for parameter reconstruction (random drift and diffusion coefficients of the Itô stochastic differential equation, SDE) in the problem of stochastic modeling of air–sea heat flux increment evolution. The first method relates to a nonparametric estimation of the transition probabilities (wherein consistency is proven). The second approach is a semiparametric reconstruction based on the approximation of the SDE solution (in terms of distributions) by finite normal mixtures using the maximum likelihood estimates of the unknown parameters. This approach does not require any additional assumptions for the coefficients, with the exception of those guaranteeing the existence of the solution to the SDE itself. It is demonstrated that the corresponding conditions hold for the analyzed data. The comparison is carried out on the simulated samples, modeling the case where the SDE random coefficients are represented in trigonometric form, which is related to common climatic models, as well as on the ERA5 reanalysis data of the sensible and latent heat fluxes in the North Atlantic for 1979–2022. It is shown that the results of these two methods are close to each other in a quantitative sense, but differ somewhat in temporal variability and spatial localization. The differences during the observed period are analyzed, and their geophysical interpretations are presented. The semiparametric approach seems promising for physics-informed machine learning models. [ABSTRACT FROM AUTHOR]

Elevators have become an integral part of modern buildings, and technological advances have enabled the monitoring of their operational status through sensor technology. In response to the development of the elevator industry and the need for practical elevator operation risk evaluation, this paper proposes an elevator risk evaluation method based on fuzzy theory and machine learning methods. The method begins by establishing an elevator operation risk evaluation index system. The traditional fuzzy comprehensive evaluation method is then employed to evaluate the risk levels of the 50 elevators studied. The collected index data and labels (fuzzy comprehensive evaluation results) are used as inputs to train the support vector machine (SVM) model. To optimize the SVM model, the maximum information coefficient method, enhanced by the correlation-based feature selection (MIC-CFS) method, is employed to select features for the index input to the SVM model. The improved gray wolf algorithm (IGWO) method optimizes the SVM. Finally, the model's performance is verified using new index data. The experimental results demonstrate that introducing machine learning methods for elevator risk evaluation saves time and effort while providing good accuracy compared to the traditional expert evaluation method. The optimization of the SVM model by IGWO and feature selection by the MIC-CFS method results in a more concise SVM model that converges faster during training, exhibits better stability, and achieves higher accuracy. [ABSTRACT FROM AUTHOR]

Short-term traffic flow forecasting, an essential enabler for intelligent transportation systems, is a fundamental and challenging task for dramatically changing traffic flow over time. In this paper, we present a gravitational search optimized kernel extreme learning machine, named GSA-KELM, to avoid manually traversing all possible parameters to improve the potential performance. Furthermore, with the interference of heavy-tailed impulse noise, the performance of KELM may be seriously deteriorated. Based on the Kalman filter that cleverly combines observed data and estimated data to perform the closed-loop management of errors and limit the errors within a certain range, we propose a combined model, termed GSA-KELM-KF. The experimental results of two real-world datasets demonstrate that GSA-KELM-KF outperforms the state-of-the-art parametric and non-parametric models. [ABSTRACT FROM AUTHOR]