Search

This work investigates the significance of Hilbert domain characterization of wavelet packets in classifying different emotion of speech signal. The goal of this paper is to create a new emotional speech database and introduce a new feature extraction approach that can recognize various emotions. The proposed feature, wavelet cepstral coefficients (WCC) are based on Hilbert spectrum analysis of the wavelet packet of the speech signal. The speaker-independent machine learning models are developed using multiclass support vector machine (SVM) and k-nearest neighbourhood (KNN) classifier. The approach is tested with newly developed Telugu Indian database and the EMOVO (Italian emotional speech) database. Our proposed wavelet features achieve a peak accuracy of 73.5%, further boosted by NCA feature selection by 3–5%, resulting in an improved unweighted average recall (UAR) of 78% for database 1 and 87.50% for database 2, employing optimal wavelet features in conjunction with SVM classification. The proposed features outperformed the baseline Mel-frequency cepstral coefficients (MFCC) feature. The performance of newly formulated features is better than other existing methodologies tested with different language databases. [ABSTRACT FROM AUTHOR]

Machine learning (ML) techniques have shown great potential in cardiovascular surgery, including real-time stenosis recognition, detection of stented coronary anomalies, and prediction of in-stent restenosis (ISR). However, estimating neointima evolution poses challenges for ML models due to limitations in manual measurements, variations in image quality, low data availability, and the difficulty of acquiring biological quantities. An effective in silico model is necessary to accurately capture the mechanisms leading to neointimal hyperplasia. Physics-informed neural networks (PINNs), a novel deep learning (DL) method, have emerged as a promising approach that integrates physical laws and measurements into modeling. PINNs have demonstrated success in solving partial differential equations (PDEs) and have been applied in various biological systems. This paper aims to develop a robust multiphysics surrogate model for ISR estimation using the physics-informed DL approach, incorporating biological constraints and drug elution effects. The model seeks to enhance prediction accuracy, provide insights into disease progression factors, and promote ISR diagnosis and treatment planning. A set of coupled advection-reaction-diffusion type PDEs is constructed to track the evolution of the influential factors associated with ISR, such as platelet-derived growth factor (PDGF), the transforming growth factor- β (TGF- β ), the extracellular matrix (ECM), the density of smooth muscle cells (SMC), and the drug concentration. The nature of PINNs allows for the integration of patient-specific data (procedure-related, clinical and genetic, etc.) into the model, improving prediction accuracy and assisting in the optimization of stent implantation parameters to mitigate risks. This research addresses the existing gap in predictive models for ISR using DL and holds the potential to enhance patient outcomes through predictive risk assessment. [ABSTRACT FROM AUTHOR]

The unsupervised anomaly detection in high-dimensional and complex settings poses a formidable challenge. To tackle the challenges associated with the recognition of high-dimensional data, this paper proposes a feedback channel between the Variational Auto Encoder and the Gaussian process to enhance its data feature extraction capabilities. In order to alleviate the impact of common mislabeling issues within unsupervised contexts, the proposed model introduces a method for the selection of representative data. Subsequently, these representative data points are employed as anchors for the construction of a full-data Gaussian process model covering the entire dataset. To further disentangle the intertwined normal and abnormal data, the proposed model employs a Bayesian framework to guide the displacement of each data point within the latent space toward its position with the highest confidence. This strategic approach effectively discriminates between normal and abnormal data. This innovative method significantly enhances data consolidation within the latent space, thus preserving robustness, particularly when processing high-complexity data. Also, the numerical experiments not only validate the method's effectiveness but also showcase the model's robustness. These experiments highlight that even a small number of incorrect labels do not significantly impact the accuracy of anomaly detection. [ABSTRACT FROM AUTHOR]

In recent years, various studies have been conducted on SVMs and their applications in different area. They have been developed significantly in many areas. SVM is one of the most robust classification and regression algorithms that plays a significant role in pattern recognition. However, SVM has not been developed significantly in some areas like large-scale datasets, unbalanced datasets, and multiclass classification. Efficient SVM training in large-scale datasets is of great importance in the big data era. However, as the number of samples increases, the time and memory required to train SVM increase, making SVM impractical even for a medium-sized problem. With the emergence of big data, this problem becomes more significant. This paper presents a novel distributed method for SVM training in which a very small subset of training samples is used for classification, which reduces the problem size and thus the required memory and computational resources. The solution of this problem almost converges to standard SVM. This method includes three steps: first, detecting a subset of distributed training samples, second, creating local models of SVM and obtaining partial vectors, and finally combining the partial vectors and obtaining the global vector and the final model. In addition, the datasets which suffer from unbalanced number of samples and tend to the majority class, the proposed method balances the samples of the two classes and it can be used in unbalanced datasets. The empirical results show that using this method is efficient for large-scale problems. [ABSTRACT FROM AUTHOR]

This review paper presents the practical development of imaging biomarkers in the scope of the PRIMAGE (PRedictive In silico Multiscale Analytics to support cancer personalized diaGnosis and prognosis, Empowered by imaging biomarkers) project, as a noninvasive and reliable way to improve the diagnosis and prognosis in pediatric oncology. The PRIMAGE project is a European multi-center research initiative that focuses on developing medical imaging-derived artificial intelligence (AI) solutions designed to enhance overall management and decision-making for two types of pediatric cancer: neuroblastoma and diffuse intrinsic pontine glioma. To allow this, the PRIMAGE project has created an open-cloud platform that combines imaging, clinical, and molecular data together with AI models developed from this data, creating a comprehensive decision support environment for clinicians managing patients with these two cancers. In order to achieve this, a standardized data processing and analysis workflow was implemented to generate robust and reliable predictions for different clinical endpoints. Magnetic resonance (MR) image harmonization and registration was performed as part of the workflow. Subsequently, an automated tool for the detection and segmentation of tumors was trained and internally validated. The Dice similarity coefficient obtained for the independent validation dataset was 0.997, indicating compatibility with the manual segmentation variability. Following this, radiomics and deep features were extracted and correlated with clinical endpoints. Finally, reproducible and relevant imaging quantitative features were integrated with clinical and molecular data to enrich both the predictive models and a set of visual analytics tools, making the PRIMAGE platform a complete clinical decision aid system. In order to ensure the advancement of research in this field and to foster engagement with the wider research community, the PRIMAGE data repository and platform are currently being integrated into the European Federation for Cancer Images (EUCAIM), which is the largest European cancer imaging research infrastructure created to date. [ABSTRACT FROM AUTHOR]

Landslide susceptibility mapping (LSM) is a major measure for disaster prevention and control. This paper proposes an incremental learning random forest model that uses random forests for incremental training of landslides. The 1652 landslides in Ganzhou, China from year 2015 to 2020 were divided into three equal parts (T1, T2 and T3) considering time series. And the data of T1 part is used to construct incremental learning random forest model ILRF_1, T2 and T3 part as the new part of incremental training. Landslide susceptibility maps were plotted using ILRF, batch learning random forests (BLRF), batch learning support vector machines (BLSVM), respectively. And the accuracy is verified using the receiver operating curve (ROC). The results showed that the ILRF was the most promising model(AUC = 0.937) for the LSM, followed by BLRF(AUC = 0.905), BLSVM(AUC = 0.865). ILRF has better predictive ability in the susceptibility evaluation of this region. It is very high-prone area accounts for the smallest proportion, mainly concentrated in the mountain front area in the northeast of Ganzhou City. The multi-year average rainfall, population density, and normalized difference vegetation index are the first three important factors that control the spatial possibility of landslides. [ABSTRACT FROM AUTHOR]

It is of great significance to accurately grasp ground-level ozone concentration on a wide scale in order to cope with the increasingly serious ground-level ozone pollution problem. Currently, satellite technology can be used to monitor ground-level ozone concentration on a wide scale, but there is the problem of insufficient spatial resolution. The use of ground monitoring stations to detect ground-level ozone concentration can obtain more accurate ozone concentration around the station, but there are problems of limited number of stations and sparse distribution. Therefore, it is still a challenge to accurately grasp the ground-level ozone concentration at a large scale and high spatial resolution. To address this issue, this paper proposes a ground-level ozone concentration inversion algorithm based on deep learning methods, incorporating data from remote sensing satellites, ground monitoring stations, meteorological conditions, geography, and land use. Taking Yunnan Province in China as the experimental region and utilizing five-fold cross-validation, the model's Mean Square Error (MSE) reached 340.45 and Root Mean Square Error (RMSE) reached 18.45. Compared to machine learning models, the error decreased by 14.53%, outperforming Random Forest models, Convolution Neural Networks, and Vision Transformers. This study offers a referential approach for accurately capturing ground-level ozone concentrations at a large scale with high spatial resolution. [ABSTRACT FROM AUTHOR]

Surface Water Mapping (SWM) is essential for studying hydrological and ecological phenomena. SWM holds significant importance in water resource management, environmental conservation, and disaster preparation. Recently, rapid urbanization, overutilization, and environmental degradation have seriously impacted surface water bodies. Rapid advancement in remote sensing data and technologies has promoted the SWM to a new era. Timely and precise SWM is crucial for water resource preservation and planning. This paper critically reviews the extraction of surface water bodies from optical sensors using Spectral Indices (SI), Machine Learning (ML), Deep Learning (DL), and Spectral unmixing with a comprehensive overview of satellite data, study areas, methodologies, results, advantages, and disadvantages, especially over the last decade. The extensive review of SWM reveals that DL outperforms ML and SI. DL outperforms other methods because it incorporates crucial elements in network design, like skip connections, dilation convolution, attention mechanisms, and residual blocks. The spectral unmixing addresses the mixed pixel misclassification problem. Some SI, ML, and DL methods are implemented, and the results are discussed. Integrating the DL technique with spectral unmixing, fusing multisource data (SAR and optical) and integrating it with ancillary data (DEM) is the future direction for improved SWM. [ABSTRACT FROM AUTHOR]

Detection of volatile organic compounds (VOCs) from the breath is becoming a viable route for the early detection of diseases non-invasively. This paper presents a sensor array of 3 component metal oxides that give maximal cross-sensitivity and can successfully use machine learning methods to identify four distinct VOCs in a mixture. The metal oxide sensor array comprises NiO-Au (ohmic), CuO-Au (Schottky), and ZnO–Au (Schottky) sensors made by the DC reactive sputtering method and having a film thickness of 80–100 nm. The NiO and CuO films have ultrafine particle sizes of < 50 nm and rough surface texture, while ZnO films consist of nanoscale platelets. This array was subjected to various VOC concentrations, including ethanol, acetone, toluene, and chloroform, one by one and in a pair/mix of gases. Thus, the response values show severe interference and departure from commonly observed power law behavior. The dataset obtained from individual gases and their mixtures were analyzed using multiple machine learning algorithms, such as Random Forest (RF), K-Nearest Neighbor (KNN), Decision Tree, Linear Regression, Logistic Regression, Naive Bayes, Linear Discriminant Analysis, Artificial Neural Network, and Support Vector Machine. KNN and RF have shown more than 99% accuracy in classifying different varying chemicals in the gas mixtures. In regression analysis, KNN has delivered the best results with an R2 value of more than 0.99 and LOD of 0.012 ppm, 0.015 ppm, 0.014 ppm, and 0.025 ppm for predicting the concentrations of acetone, toluene, ethanol, and chloroform, respectively, in complex mixtures. Therefore, it is demonstrated that the array utilizing the provided algorithms can classify and predict the concentrations of the four gases simultaneously for disease diagnosis and treatment monitoring. [ABSTRACT FROM AUTHOR]

The primary objective of this paper is to develop an appropriate predictive formula for the compressive strength of pervious concrete, which depends on its mixture. This will allow for the improvement of the proportioning procedure that considers both the target porosity and target compressive strength. To achieve this, an effective computational strategy is first constructed and investigated for the creation of simple and easily applicable symbolic regression functions within the Genetic Programming-based Symbolic Regression framework. Recent advancements in fast logical parallelism and model-based algorithms are also applied to perform calculations on a large quantity of examples, with the aim of finding the most suitable analytical solutions at a low computational cost. Next, to assess the effectiveness of this model in predicting the compressive strength of concrete in general, computations are carried out using the well-known Yeh's dataset on conventional concrete compressive strength. This dataset has been extensively studied using both "black-box" and "white-box" machine learning algorithms. The results reveal that more suitable formulas can be generated through this computational process, compared to several scenarios discussed in the literature. Furthermore, the model is extended to pervious concrete, based on the dataset of 164 samples of 28-day compressive strength collected from 14 different sources. The findings for pervious concrete exhibited high accuracy compared to the most effective black-box models and micromechanical/empirical models, with a coefficient of determination of approximately 0.9 for simple predictive equations, thereby supporting the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Deep learning methods have shown significant potential in computer vision applications within the field of civil engineering. One important area of research is the detection and classification of construction machines using convolutional neural networks (CNNs). However, a major challenge in adopting machine learning models for vision systems lies in hyperparameter optimization. Therefore, the objective of this paper is to propose a meticulous statistical approach to hyperparameter tuning for deep learning in the visual classification of construction machines. It is noteworthy that the method for selecting hyperparameters utilizes statistical concepts, including analysis of variance and the Tukey test. Moreover, three research questions were formulated to address statistical differences in performance when varying the CNN architecture, optimizer and configuration (architecture and optimizer). Two additional research questions were posed to determine the recommended hyperparameters for the task and assess performance differences for non-selected hyperparameters. For this purpose, 18 hyperparameter combinations were analyzed, encompassing three CNN architectures (DenseNet, Mobilenet and VGG16) and six optimizers (adadelta, adagrad, rmsprop, adam, adamax and sgd). The Alberta Construction Image Dataset, consisting of 2850 images of three construction machines (excavator, dump truck and concrete mixer truck), was used in the experiments. The results demonstrate that the hyperparameter configurations exhibit statistically significant differences in image classification. Furthermore, the recommended combination (DenseNet × adagrad) yielded the highest average accuracy results in both binary tests (90.0%) and multiclass classification (77.8%). [ABSTRACT FROM AUTHOR]

Complete anomaly detection strategies that are both signal sensitive and compatible with background estimation have largely focused on resonant signals. Non-resonant new physics scenarios are relatively under-explored and may arise from off-shell effects or final states with significant missing energy. In this paper, we extend a class of weakly supervised anomaly detection strategies developed for resonant physics to the non-resonant case. Machine learning models are trained to reweight, generate, or morph the background, extrapolated from a control region. A classifier is then trained in a signal region to distinguish the estimated background from the data. The new methods are demonstrated using a semi-visible jet signature as a benchmark signal model, and are shown to automatically identify the anomalous events without specifying the signal ahead of time. [ABSTRACT FROM AUTHOR]

Lithium-ion batteries are becoming more popular due to their superior performance like high power density, long lifespan, broad operating range of temperatures, quick charging capabilities, and low self-discharge. The implementation of a Battery Management System (BMS) is crucial in order to guarantee the secure and optimal functioning of electric vehicle batteries. BMS monitors, controls, and maintains the health and performance of rechargeable batteries by monitoring the State of Charge (SoC), State of Health (SoH), and State of Temperature (SoT). SoC quantifies the amount of energy stored in a battery at a certain moment, and it is employed to approximate the remaining distance that may be covered. SoH describe the overall health or state of a rechargeable battery. It reveals how well a battery performs in comparison to its original state. SoT estimation keeps the battery temperature within a safe range, allowing it to reach a higher age and safety, which is critical for battery reliability. In this paper SoC, SoH, SoT estimation models for a lithium-ion battery have been developed using an improved EP-based R110-BLSTM approach. The Emperor Penguin based Residual Network-110 incorporated Bidirectional Long-Short Term Memory (EP-based R110-BLSTM) is ideal for estimating SoC, SoH, and SoT and features great accuracy, a quick estimation speed, and strong generalization capabilities. However, Extreme Learning Machine performance is heavily reliant on proper feature extraction. In order to enhance estimate performance by extracting the best features, Simulated-Annealing-based Golden Eagle optimization is utilized. Electric vehicle drive cycles are used to test the model's resilience to temperature changes. MATLAB 2018b software is used in the execution of this research. The results demonstrate that the proposed model performs better in terms of accuracy and has lower SoC, SoH, and SoT error rates than existing models. A thorough comparison between the recommended model and existing methods is also made, further demonstrating the proposed model's superiority. [ABSTRACT FROM AUTHOR]

Gaussian process (GP) regression provides a probabilistic framework for modeling geochemistry in mineral resource estimation and environmental monitoring applications. An issue with this approach is that the kernel hyperparameters obtained by maximizing the log-marginal likelihood (LML) often produce GP posterior mean estimates that are overly smooth. This motivates the development of augmented kernels that are more capable of capturing the variability inherent in geological/geochemical processes than the existing stationary covariance functions like the Matérn kernels. This paper makes two contributions. First, it describes an extended class of stationary kernels that contain an extra smoothness hyperparameter (α ) which can be learned from the input data. Valid intervals for α that lead to positive semi-definiteness are determined. Second, it uses a statistical measure called the structural similarity index (SSIM) to quantify smoothness and the spatial fidelity of GP solutions with respect to the input samples. This provides a new way for validating and optimizing GP models. Statistical and spectral analyses provide insights into the behavior of α in the augmented kernels which retain useful properties such as sparsity. Results on the Northern Great Basin geochemical dataset demonstrate that, all things being equal, (1) adjusting α increases spatial fidelity in GP regression; (2) SSIM is a more reliable spatial quality measure than LML; and (3) the optimal α value obtained is correlated with the Wiener entropy of the random process, which indicates the spectral flatness of the chemical signal in the Fourier domain. For GP regression, function smoothness may be defined using Sobolev space. The results show that α regulates over-smoothing by moderating the rate of decay in the power spectrum of the equivalent kernel. [ABSTRACT FROM AUTHOR]

With the rapid advancements of sensor technologies and mobile computing, Mobile Crowd Sensing (MCS) has emerged as a new paradigm to collect massive-scale rich trajectory data. Nomadic sensors empower people and objects with the capability of reporting and sharing observations on their state, their behavior and/or their surrounding environments. Processing and mining multi-source sensor data in MCS raise several challenges due to their multi-dimensional nature where the measured parameters (i.e., dimensions) may differ in terms of quality, variability, and time scale. We consider the context of air quality MCS and focus on the task of mining the micro-environment from the MCS data. Relating the measures to their micro-environment is crucial to interpret them and analyse the participant's exposure properly. In this paper, we focus on the problem of investigating the feasibility of recognizing the human's micro-environment in an environmental MCS scenario. We propose a novel approach for learning and predicting the micro-environment of users from their trajectories enriched with environmental data represented as multidimensional time series plus GPS tracks. We put forward a multi-view learning approach that we adapt to our context, and implement it along with other time series classification approaches. We extend the proposed approach to a hybrid method that employs trajectory segmentation to bring the best of both methods. We optimise the proposed approaches either by analysing the exact geolocation (which is privacy invasive), or simply applying some a priori rules (which is privacy friendly). The experimental results, applied to real MCS data, not only confirm the power of MCS and air quality (AQ) data in characterizing the micro-environment, but also show a moderate impact of the integration of mobility data in this recognition. Furthermore, and during the training phase, multi-view learning shows similar performance as the reference deep learning algorithm, without requiring specific hardware. However, during the application of models on new data, the deep learning algorithm fails to outperform our proposed models. [ABSTRACT FROM AUTHOR]

Predictive analytics over mobility data is of great importance since it can assist an analyst to predict events, such as collisions, encounters, traffic jams, etc. A typical example is anticipated location prediction, where the goal is to predict the future location of a moving object, given a look-ahead time. What is even more challenging is to be able to accurately predict collective behavioural patterns of movement, such as co-movement patterns as well as their course over time. In this paper, we address the problem of Online Prediction of Co-movement Patterns. Furthermore, in order to be able to calculate the accuracy of our solution, we propose a co-movement pattern similarity measure, which facilitates the comparison between the predicted clusters and the actual ones. Finally, we calculate the clusters' evolution through time (survive, split, etc.) and compare the cluster evolution predicted by our framework with the actual one. Our experimental study uses two real-world mobility datasets from the maritime and urban domain, respectively, and demonstrates the effectiveness of the proposed framework. [ABSTRACT FROM AUTHOR]

With the increasing number of ransomware attacks on critical infrastructures, there is an urgent need to develop effective systems that can detect ransomware early. In order to achieve this objective, many detection solutions rely on machine learning to analyze the features of ransomware samples. However, these solutions often need to execute ransomware to extract sufficient features, which can increase the risk of virus infection. This paper proposes a novel static analysis framework based on the portable executable header, which utilizes deep reinforcement learning for early detection of ransomware. The framework leverages the portable executable header (PE header) from executable files as the key feature to identify ransomware. Moreover, the framework can learn the features of ransomware by using deep reinforcement learning. This is achieved through the interaction between the agent and the environment, and then the samples are segmented into ransomware and benign categories by taking actions. The proposed framework achieves rapid detection speeds without running ransomware samples by employing a lightweight network and the portable executable header from the raw byte files. To the best of our knowledge, we are the first to exploit deep reinforcement learning on the PE header for ransomware early detection. Our experiments on two real-world datasets reveal that the proposed framework can rapidly detect unseen ransomware. [ABSTRACT FROM AUTHOR]

The advanced development of communication technologies has made covert communications quite challenging to be recognized. By altering an entity's timing behavior during overt network communication, a covert timing channel (CTC) provides a way to leak sensitive data. This is becoming a serious threat due to the ability to transmit hidden messages without being detected by traditional security systems such as proxies and firewalls. In this paper, we explore the automatic classification and identification of covert timing channels using deep neural networks, namely, Long Short-Term Memory (LSTM), 1D-Convolutional Neural networks (1D-CNN), and a hybrid of LSTM and 1D-CNN. These classifiers have been trained and tested using sequence real inter-arrival times datasets labeled with covert or legitimate. The stream length differs for each dataset; (32, 64, 128, 256, and 512 bytes). Traditional machine-learning models have also been developed for comparisons and evaluation. The evaluation results showed that the hybrid of LSTM and CNN model outperforms other models either developed by deep neural networks or traditional machine learning with an accuracy of %97.5. [ABSTRACT FROM AUTHOR]

Organizations widely use cloud computing to outsource their computing needs. One crucial issue of cloud computing is that services must be available to clients at all times. However, the cloud services may be temporarily unavailable due to maintenance of the cloud infrastructure, load balancing of services, defense against cyber attacks, power management, proactive fault tolerance, or resource usage. The unavailability of cloud services impacts negatively on the business model of cloud providers. One solution to tackle the service unavailability is Live Virtual Machine Migration (LVM), that is, moving virtual machines (VMs) from the source host machine to the destination host without disrupting the running application. Pre-copy memory migration is a common LVM approach used in most networked systems such as the cloud. The main difficulty with this approach is the high rate of frequently updating memory pages, referred to as "dirty pages. Transferring these updated or dirty pages during the pre-copy migration approach prolongs the total migration time. After a predefined iteration, the pre-copy approach enters the stop-and-copy phase and transfers the remaining memory pages. If the remaining pages are huge, the downtime or service unavailability will be very high -resulting in a negative impact on the availability of the running services. To minimize such service downtime, it is critical to find an optimal time to migrate a virtual machine in the pre-copy approach. To address the issue, this paper proposes a machine learning-based method to optimize pre-copy migration. It has mainly three stages (i) Feature selection (ii) Model generation and (iii) Application of the proposed model in pre-copy migration. The experiment results show that our proposed model outperforms other machine learning models in terms of prediction accuracy and it significantly reduces downtime or service unavailability during the migration process. [ABSTRACT FROM AUTHOR]

The sandy conglomerate reservoir in layer Es3 of the Liaohe Eastern Depression has good potential for oil reservoir exploration and has been identified as a key area for future exploration. The low porosity and permeability, complex lithology, and strong heterogeneity of the target layer make it difficult to predict favorable reservoirs. The objective of this study is to analyze and process conventional logging data to extract feature parameters that affect lithology by establishing a decision tree lithology classifier. Principal component analysis is used to reduce data dimensionality, and the elbow method is applied to the clustering algorithm to establish the optimal number of clusters for the automatic classification of reservoir types. Further, support vector machines are used for lithology classification based on features with higher classification capabilities. The results show that the support vector machine lithology recognition method based on feature selection achieved an accuracy of 91.8%. The processing of actual well data has verified the feasibility of the method. Based on the combination of core experiments and oil testing results, the characteristics of three types of reservoirs were presented, and potential reservoir zones were proposed for drilling wells. The comprehensive analysis and the practical application of the developed method reveal that the class I reservoir has high hydrocarbon production and could be the most favorable reservoir in the Es3 sandy conglomerate. The processing data of lithology identification and reservoir classification evaluation are consistent with core data and hydrocarbon production data, verifying the effectiveness and practicability of the method proposed in this paper. The results of this study will serve as a reference for low porosity and permeability sandy conglomerate reservoir evaluation based on machine learning in the target area. [ABSTRACT FROM AUTHOR]

The concept of geo-localization broadly refers to the process of determining an entity's geographical location, typically in the form of Global Positioning System (GPS) coordinates. The entity of interest may be an image, a sequence of images, a video, a satellite image, or even objects visible within the image. Recently, massive datasets of GPS-tagged media have become available due to smartphones and the internet, and deep learning has risen to prominence and enhanced the performance capabilities of machine learning models. These developments have enabled the rise of image and object geo-localization, which has impacted a wide range of applications such as augmented reality, robotics, self-driving vehicles, road maintenance, and 3D reconstruction. This paper provides a comprehensive survey of visual geo-localization, which may involve either determining the location at which an image has been captured (image geo-localization) or geolocating objects within an image (object geo-localization). We will provide an in-depth study of visual geo-localization including a summary of popular algorithms, a description of proposed datasets, and an analysis of performance results to illustrate the current state of the field. [ABSTRACT FROM AUTHOR]

This paper investigates the capability of plain Vision Transformers (ViTs) for semantic segmentation using the encoder–decoder framework and introduce SegViTv2. In this study, we introduce a novel Attention-to-Mask (ATM) module to design a lightweight decoder effective for plain ViT. The proposed ATM converts the global attention map into semantic masks for high-quality segmentation results. Our decoder outperforms popular decoder UPerNet using various ViT backbones while consuming only about 5 % of the computational cost. For the encoder, we address the concern of the relatively high computational cost in the ViT-based encoders and propose a Shrunk++ structure that incorporates edge-aware query-based down-sampling (EQD) and query-based up-sampling (QU) modules. The Shrunk++ structure reduces the computational cost of the encoder by up to 50 % while maintaining competitive performance. Furthermore, we propose to adapt SegViT for continual semantic segmentation, demonstrating nearly zero forgetting of previously learned knowledge. Experiments show that our proposed SegViTv2 surpasses recent segmentation methods on three popular benchmarks including ADE20k, COCO-Stuff-10k and PASCAL-Context datasets. The code is available through the following link: https://github.com/zbwxp/SegVit. [ABSTRACT FROM AUTHOR]

Solar radio bursts are intense radio radiation sources that occur during the energy-release process and represent a hot topic in solar-physics and space-weather research. In this paper, we present a multimode prediction model for daily solar radio bursts. The model uses deep learning and machine learning to obtain data information from different dimensions and to establish the relationship between the characteristics of the solar active region on the solar surface and solar radio bursts. For this model, we use data from the Solar and Heliospheric Observatory (SOHO)/Michelson Doppler Imager (MDI) total solar magnetic map, the Royal Observatory of Belgium World Data Centre in Brussels, and NOAA sunspot parameters (including number, area, and type of sunspots) as inputs. The output results are then compared with the list of solar radio bursts recorded by the Radio Solar Telescope Network (RSTN) to determine whether solar radio bursts are present and to determine the key parameters for determining radio bursts. Based on 5449 days of observational data, we find that the prediction accuracy of the model is 0.898 ± 0.011, and that the number of sunspots is a key parameter in determining the occurrence of solar radio bursts. Specifically, when the number of sunspots is greater than 15, the probability of occurrence of solar radio bursts is greater than 90%. We have identified the key parameters and thresholds for determining solar radio bursts and highlighted the key parameters for space-weather prediction. In addition, the prediction model can also be used for predicting in other fields. [ABSTRACT FROM AUTHOR]

With the development of informatization, traditional industries have been seriously impacted by cross-border e-commerce (CBEC), and the financial management mode no longer meets the needs of CBEC users. As cross-border electronic trading platforms face enormous financial data challenges, they need financial supervision and effective network risk protection. Artificial intelligence (AI) can effectively improve the financial accounting ability and resource integration of platforms. However, at present, the underdeveloped financial management system of CBEC platforms and problems, such as the running records of platform funds have seriously hindered the development of the financial standardization of platforms because the financial management system affects managers' statistical analysis of financial data. Therefore, this paper analyzes the financial risks, the management problems and the causes of these problems of CBEC platforms and then uses AI to study the financial operation of CBEC platforms. Subsequently, this paper uses a machine learning (ML) algorithm to analyze the financial data clustering center and the security factor of the financial model. Finally, this paper proposes some corresponding strategies for financial model optimization and construction that can improve the information security and capital management of CBEC finance and promote the long-term development of CBEC platforms. The experimental results show that the classifier value and the safety factor of the financial model of CBEC platforms gradually increase under the ML algorithm. The mean value of the classifier value is approximately 1.14, and the mean value of the safety factor is approximately 1.37. Overall, the initial value of the classifier value of the platform financial model is 0.85, which increases to 1.41 on the seventh day, while the whole process increases by 0.56. The initial value of the safety factor of the platform financial model is 1.10, which increases to 1.64 on the seventh day, while the whole process increases by 0.54. The transaction information security and financial accounting accuracy of the CBEC platform financial model is better than that of the original financial model. The transaction information security is 10.4% higher than that of the original financial model, and the financial accounting accuracy is 10% higher than that of the original financial model. In other words, both AI and ML can promote the financial accounting standards and long-term development of CBEC platforms. [ABSTRACT FROM AUTHOR]

Nowadays many tools are in use in processes related to handling bug reports, feature requests, supporting questions or similar related issues which should be handled during software development or maintenance. Part of them use machine learning techniques. In introduction is presented a review of fundamental methods used for evaluation of machine learning models. This paper points out weak points of currently used metrics for evaluation in specific context of the cases related to software development especially bug reports. The disadvantages of state of the art are related to disregarding time dependencies which are important to be applied for creating train and test sets as they may have impact on results. Extensive research of the art has been conducted and has not been found any article with the use of time dependencies for evaluation of machine learning models in the context of works related to software development applications like machine learning solutions to supporting bug tracking systems. This paper introduces a novel solution which is devoid of these drawbacks. Experimental research showed the effectiveness of the introduced method and significantly different results obtained compared to the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Model-agnostic meta-learning (MAML) highlights the ability to quickly adapt to new tasks with only a small amount of labeled training data among many few-shot learning algorithms. However, the computational complexity is high, because the MAML algorithm generates a large number of second-order parameters in the secondary gradient update. In addition, due to the non-convex nature of the neural network, the loss landscape has many flat areas, leading to slow convergence during training, and excessively long training. In this paper, a second-order optimization method called Kronecker-factored Approximate Curvature (K-FAC) is proposed to approximate Natural Gradient Descent. K-FAC reduces the computational complexity by approximating the large matrix of the Fisher information as the Kronecker product of two much smaller matrices, and the second-order parameter information is fully utilized to accelerate the convergence. Moreover, in order to solve the problem that Natural Gradient Descent is sensitive to the learning rate, this paper proposes Kronecker-factored Approximate Curvature with adaptive learning rate for optimizing model-agnostic meta-learning (AK-MAML), which automatically adjusts the learning rate according to the curvature and improves the efficiency of training. Experimental results show that AK-MAML has the ability of faster convergence, lower computation, and higher accuracy on few-shot datasets. [ABSTRACT FROM AUTHOR]

The issues of classification and characterization of surface operational defects according to the results of ultrasonic, eddy current, and visual inspection methods of nondestructive testing are considered. At the same time, the visual inspection method was realized with the use of a television inspection camera equipped with a computer vision function and a laser triangulation sensor. The paper presents a dataset containing 5760 images of pipelines with and without pitting corrosion. A convolutional neural network (CNN) is presented that was applied to classify the images obtained from the TV inspection camera into images without corrosion and images with pitting corrosion. The paper presents a dataset containing 269 measurements of planar and volumetric surface defects. A model for surface defect sizing based on gradient boosting is presented. The paper develops an algorithm for classification and sizing of surface defects in complex diagnostics in which the obtained models are applied, and determines the accuracy of this algorithm in the RMSE metric, which was calculated within the studied test dataset and amounted to 0.011 mm. [ABSTRACT FROM AUTHOR]

Renewable energy resources are the only solution to the energy crisis over the world. Production of energy by the solar panel cells are identified as the main renewable energy resources. The generation of energy by the solar panels is affected by the cracks on it. Hence, the detection of cracks is important to increase the energy levels produced by the solar cells. In this paper, the solar panel images are classified into either cracked image or non-cracked image using deep learning algorithm. The proposed method is designed with the following modules preprocessing, enhancement, feature computations, classification and crack segmentation. The source solar panel image are denoised using adaptive median filter as the preprocessing process and then the pixels in denoised solar panel image are enhanced using cumulative enhancement (CE) method. The external features are computed from the CE enhanced solar panel image and these features are classified by Improved AlexNet (IAN)-deep learning classifier to produce the classification results as either cracked or non-cracked solar panel image. Finally, the cracks in classified cracked solar panel image are segmented using morphological algorithm. The main significance of this paper is that the proposed methods stated here detect single and multiple level of minor cracks, which enhances the energy levels production. [ABSTRACT FROM AUTHOR]

A distributed computing model for channel bandwidth allocation and optimization can involve multiple components working together to efficiently allocate and optimize the available bandwidth in a distributed system. The efficient allocation of channel bandwidth in the distributed computing model is crucial for optimizing resource utilization and improving system performance. This paper, proposed the Imperialist Competitive Spline Interpolation (ICSI) scheme, which combines computational intelligence and deep learning techniques to address the challenge of channel bandwidth allocation. The ICSI scheme optimizes resource allocation by considering user requirements and resource availability, utilizing polynomial equations and spline interpolation. The Imperialist Competitive Optimization model evaluates and optimizes the available resources in the distributed environment. With the optimized resources spline interpolation is implemented for the computation of the available resources. Extensive simulations and performance analysis demonstrate the effectiveness of the ICSI scheme in terms of resource utilization, throughput, latency, fairness index, and energy efficiency. The ICSI model achieves the minimal waiting time of 3 ms and minimal latency of 6.4 m. Comparative analysis of the Round Robin scheme further confirms the superiority of the ICSI scheme in terms of task scheduling efficiency. The findings of this paper contribute to the advancement of distributed computing models for channel bandwidth allocation, offering a promising solution for optimizing resource allocation and improving system performance in modern computing environments. [ABSTRACT FROM AUTHOR]

Machine learning (ML) is increasingly used to enhance production systems and meet the requirements of a rapidly evolving manufacturing environment. Compared to larger companies, however, small- and medium-sized enterprises (SMEs) lack in terms of resources, available data and skills, which impedes the potential adoption of analytics solutions. This paper proposes a preliminary yet general approach to identify low-cost analytics solutions for manufacturing SMEs, with particular emphasis on ML. The initial studies seem to suggest that, contrarily to what is usually thought at first glance, SMEs seldom need digital solutions that use advanced ML algorithms which require extensive data preparation, laborious parameter tuning and a comprehensive understanding of the underlying problem. If an analytics solution does require learning capabilities, a 'simple solution', which we will characterise in this paper, should be sufficient. [ABSTRACT FROM AUTHOR]

We review the measurement methods and thickness characterization algorithms of semiconductor multilayer devices. Today’s ultrahigh-density, high-energy-efficient three-dimensional semiconductor devices require an iterative semiconductor layer-stacking process. Accurate determination of nanometer-scale layer thickness is crucial for reliable semiconductor device fabrication. In this paper, we first review the commonly used semiconductor multilayer thickness measurement methods, including destructive and nondestructive measurement methods. Next, we review two approaches for thickness characterization: model-based algorithms using a physical interpretation of multilayer structures and a method using data-driven machine learning. With the growing importance of semiconductor multilayer devices, we anticipate that this study will help in selecting the most appropriate method for multilayer thickness characterization.Highlights: This paper covers various measurement methods and algorithms to detect the thickness of semiconductor multilayer devices. Model-based algorithm is suitable for cases where the number of layers is small or physical interpretation is required. Machine learning performs well for thickness characterization of multilayer devices without complex physical analysis. [ABSTRACT FROM AUTHOR]

One of the network performance challenges of Vehicular Ad Hoc Networks (VANET) is the Black Hole attack. This destructive attack severely damages the network performance if succeeded. By replying to a route request with the minimum routing cost and the maximum sequence number (SN) value, the malicious vehicle can fool the source vehicle that it has the "freshest" and best cost-effective route to the destination vehicle. As a result, all data packets intended to a legitimate destination are caught and destroyed by the malicious vehicle. This may also cause a serious risk for autonomous vehicle control system or traffic warning applications. Previous published malicious detection algorithms suffer high error rates when the malicious node uses a "close-to-legitimate" SN value to attack or actively change the information in the route reply packet. This paper proposes a Black Hole Attack Detection Algorithm (BADA) based on a machine learning approach. In VANETs, it is a challenge to differentiate the behavior of malicious vehicles as they try to avoid detection by imitating that of the normal vehicles. The BADA classification algorithm outperforms existing solutions because it relies on the history of each vehicle's route request and response behavior, and it deploys the k-Nearest Neighbors machine learning algorithm to identify malicious vehicles. The paper also proposes a Black hole Attack Detection Routing Protocol which utilizes the proposed BADA solution to realize a more secured and improved AODV-based protocol. Using the NS2 simulation system, the paper evaluates the performance of the proposed protocol and compares it with related solutions on the traffic network model in Ho Chi Minh city, Vietnam under black hole attacks using minimum SN values. Simulation results have shown that the proposed algorithm can correctly detect the malicious node higher than 99.0%, outperforming previously published algorithms. [ABSTRACT FROM AUTHOR]

Broad Learning System (BLS) is proposed as an alternative to deep learning. It has a fast adaptive model selection and online incremental learning capability, which has been successfully applied in many fields. In this paper, the BLS model is introduced into intrusion detection, and considering the weakness of the BLS model in mining the internal structural information of samples, this paper proposes a Manifold Regularized Broad Learning System based on LU decomposition (LU-MRBLS) intrusion detection. Based on the manifold hypothesis, the LU-MRBLS model firstly constructs the graph Laplacian operator in the data input space to mine the potential information of the data. Then, under the manifold regularized framework, the feature nodes, enhancement nodes, and Laplacian matrix are combined to construct the objective function to regularize and optimize the BLS model to avoid the model falling into local optimization. Finally, the LU decomposition method is used to solve the output weight matrix of the MRBLS model, shorten the training time of the MRBLS model, avoid singular value problems of the solution process, and improve the intrusion detection performance of the model. In this paper, we use the KDD Cup99 dataset for parameter selection and apply it to other network models. Through rigorous experiments, the LU-MRBLS model is applied to KDD Cup99, NSL-KDD, UNSW-NB15, and CIDDS-001 datasets with better detection results than the classical machine learning models and the latest intrusion detection models. [ABSTRACT FROM AUTHOR]

Since 2007, methods that utilize side-channel data to detect hardware Trojan (HT) problems have been widely studied. Machine learning methods are widely used for hardware Trojan detection, but with the development of integrated circuits (ICs), better results are usually obtained using deep learning methods. In this paper, we propose an architecture inspired by Residual-Block and Dense-Block and combine it with SE Attention Mechanism, which we named the Res-Dense-SE-Net network. By combining residual connectivity, dense connectivity, and attention mechanism, the Res-Dense-SE-Net network can enjoy the advantages of these three network architectures at the same time, which can improve the expressiveness and performance of the model. The Res-Dense-SE-Net network can capture the key features in the image better, and it can solve the problems of gradient vanishing and feature transfer efficiently, which can in turn improve the classification accuracy and the generalization ability of the model. Based on the publicly available AES series of hardware Trojans from TrustHub and the publicly available hardware Trojan-side channel data by Faezi et al., we evaluate the effectiveness of the method proposed in this paper. The experimental results show that when a single Trojan exists, the method proposed in this paper has a high accuracy rate; and when multiple types of hardware Trojans exist at the same time and need to be categorized, the categories of hardware Trojans can also be effectively identified, and the categorization accuracy is high compared with the existing deep learning methods. [ABSTRACT FROM AUTHOR]

The Poisson multinomial distribution (PMD) describes the distribution of the sum of n independent but non-identically distributed random vectors, in which each random vector is of length m with 0/1 valued elements and only one of its elements can take value 1 with a certain probability. Those probabilities are different for the m elements across the n random vectors, and form an n × m matrix with row sum equals to 1. We call this n × m matrix the success probability matrix (SPM). Each SPM uniquely defines a PMD . The PMD is useful in many areas such as, voting theory, ecological inference, and machine learning. The distribution functions of PMD , however, are usually difficult to compute and there is no efficient algorithm available for computing it. In this paper, we develop efficient methods to compute the probability mass function (pmf) for the PMD using multivariate Fourier transform, normal approximation, and simulations. We study the accuracy and efficiency of those methods and give recommendations for which methods to use under various scenarios. We also illustrate the use of the PMD via three applications, namely, in ecological inference, uncertainty quantification in classification, and voting probability calculation. We build an R package that implements the proposed methods, and illustrate the package with examples. This paper has online supplementary materials. [ABSTRACT FROM AUTHOR]

Considering that serious hand function damage will greatly affect the daily life of patients, its recovery mainly depends on the regular inspection and manual training of medical staff, and medical monitoring based on bioelectric signals can largely replace manual re-examination as autonomous rehabilitation technology. So, for the rationality of feature selection and the diversity of classifier design in the gesture recognition process based on electromyography (EMG) signals, this paper proposes a hand medical monitoring system based on feature selection method of feature subset average recognition rate and optimal machine learning algorithm selection, which mainly depends on the prediction of hand movement. At the same time, since most experiments are conducted in different non-public proprietary databases, the comparison between various gesture recognition methods can only be analyzed to a certain extent. Therefore, this paper uses the DB1 dataset in the large publicly available NinaPro database and combines with presently well-known 11 time-domain (TD) features and 5 frequency domain (FD) features, then uses the support vector machine (SVM) classifier to comparative analysis total 136 feature combinations under various feature numbers. Under the premise of ensuring the overall recognition rate of electromyography gesture, this method will be able to reduce the number of features in feature set, according to the change of the average remove redundant features, and construct an optimal reduced EMG feature set. Finally, through the four common hand motion classifiers based on machine learning: SVM, back propagation neural network, linear discriminant analysis, and K-nearest neighbor, this paper tests and verifies the separability of the optimal reduced EMG feature set, and based on this, selects the optimal hand motion classifier to build the optimal hand motion recognition system, improve the hand medical monitoring system, and provide technical reference for the construction of real-time medical monitoring system. [ABSTRACT FROM AUTHOR]

Zircon is the most important accessory mineral in geological research, and it records information on isotopes and trace elements, which is of great significance in Earth science research. Trace elements in zircons can be used to analyze the genesis of zircons, calculate the magma temperature and oxygen fugacity, and trace the magma source. Due to the limitation of visual dimensions, zircon information is mainly shown by low-dimensional diagrams in present studies, so high-dimensional relationships are difficult to determine during trace element analysis of zircons. However, with the development of machine learning, mining the high-dimensional relationships during trace element analysis of zircons has become possible. In this paper, four supervised learning algorithms including random forest, support vector machine, decision tree, and eXtreme Gradient Boosting, were implemented to analyze the trace elements of 3907 magmatic zircons from the GEOROC database, and a precise 13-dimensional data classifier model was established to distinguish the volcanic rift, ocean island, and convergent margin tectonic settings. Based on the results of accuracy, precision, recall, and F1 score, eXtreme Gradient Boosting is the best machine learning approach in this paper and its results for accuracy, precision, recall, and F1 score are 0.906, 0.907, 0.907, and 0.905, respectively. In summary, eXtreme Gradient Boosting could provide a high-dimensional discriminative approach to distinguish tectonic settings. [ABSTRACT FROM AUTHOR]

In recent years, with the gradual depletion of shallow mineral resources, the exploitation of deep mineral resources has become an inevitable trend. Microseismic monitoring is one of the main methods to solve high stress concentration problems such as rockbursts, roof caving and water inrush in deep mine. An accurate and fast microseismic location method is the basis of microseismic monitoring. At present, machine learning (ML) has become an important auxiliary method in the field of microseismic monitoring due to its powerful feature expression ability. Compared with other microseismic positioning methods, ML can construct a more objective positioning model. This paper uses three ML models including deep neural network (DNN), random forest (RF),and support vector regression (SVR) to construct a microseismic position method to search for the microseismic source. The travel time difference between stations is used as the input of the ML model. Since the number of field data sets is not enough to complete the training of the model, this paper uses a synthetic data set with a specific speed model as the training set and uses the field data set for testing. In order to analyze the location accuracy, we compare them with three traditional microseismic location methods. To improve the positioning performance of ML models, this paper uses a Bayesian optimizer (BO) to adjust their hyperparameters. The experimental results show that the performance of the ML model adjusted by the BO has been significantly improved. The positioning accuracy order of the three ML models is DNN > RF > SVR > traditional location method. The average positioning accuracies of the DNN inside and outside the sensors array are 27.81 m and 145.96 m, respectively. For the model proposed in this paper, the positioning accuracy inside the sensors array is significantly higher than that outside the array, which is similar to the traditional positioning method. In addition, the model has a certain tolerance to the error of the speed model. Highlights: Utilization of machine learning (ML) for accurate and objective microseismic source location. Comparative analysis of three ML models (DNN, RF, SVR) and traditional methods. Enhancement of ML model performance using Bayesian optimizer for hyperparameter tuning. Significant improvement in positioning accuracy with DNN model inside and outside the sensor array. Tolerance to speed model errors in the proposed ML-based microseismic positioning method. [ABSTRACT FROM AUTHOR]

Code smells violate software development principles that make the software more prone to errors and changes. Researchers have developed code smell detectors using manual and semi-automatic methods to identify these issues. However, three key challenges have limited the practical use of these detectors: developers' subjective perceptions of code smells, lack of consensus in the detection process, and difficulty in setting appropriate detection thresholds. While code smell detection using machine learning has progressed significantly, there still appears to be a gap in understanding the effective utilization of deep learning (DL) approaches. This paper aims to review and identify current methods for code smell detection using DL techniques. A systematic literature review is conducted on 35 primary studies from a collection of 8739 publications between 2013 and the present. The analysis reveals that common code smells detected include Feature Envy, God Classes, Long Methods, Complex Classes, and Large Classes. The most popular DL algorithms used are Recurrent Neural Networks (RNN) and Convolutional Neural Networks (CNN), often combined with other techniques for better results. Algorithms that train models on large datasets with fewer independent variables demonstrate exemplary performance. The paper also highlights open issues and provides guidelines for future metric identification and selection research. [ABSTRACT FROM AUTHOR]

Technological advancement has led to a dynamic change in how people interact. The interaction can be classified as—social networks (virtual) and real-life (physical) interaction. The relationships among nodes of social networks depend on communication among nodes and have been established by transmitting messages publicly or privately, depending on the trust among nodes. However, it poses three main challenges: (i) nodes can reveal their private message, (ii) private conversation among nodes can be hacked, and (iii) stored messages can be sent to another node, therefore, compromising privacy. The paper presents a privacy-preserving model using two major techniques, Binomial Distribution and Fuzzy Logic. This process has been followed by building predictive models using machine learning to provide potential directions to mitigate security threats posed to social networking platforms. The paper also provides a wholesome discussion of model design for predictive modelling, whereas most state-of-the-art techniques were primarily concerned about delivering potential directions for predictive modelling. The employed statistical and predictive modelling techniques suggest that friendship relationships in real life among people would transcend to ensure a secure and privacy-preserving relationship among them on social networks. Empirical results indicated that XG Boost supersedes the state-of-the-art models by achieving a predictive accuracy of 92.035%. In contrast, introducing new statistical features improves the performance of all our predictive models. [ABSTRACT FROM AUTHOR]

In recent years, the rapid advancement of artificial intelligence (AI) has revolutionized various industries, with image processing playing an important role in the revolution. AI-powered image analysis has opened up new possibilities in healthcare diagnostics, autonomous navigation, and security surveillance, fuelling demand for sophisticated image-processing solutions. This paper describes an innovative approach to AI image processing that combines multi-sensor data fusion with convolutional neural networks (CNNs) within a fuzzy neural network framework. This integration uses data from various sensors, including cameras, lidar, and radar, to improve the robustness and precision of image analysis and interpretation. The T–S model serves as the foundation for the information fusion strategy. A comprehensive investigation of deep learning algorithms reveals inherent strengths such as robustness and parallelism. However, it also identifies limitations, particularly in image segmentation tasks, characterized by challenges like premature convergence and prolonged computation times. The paper proposes a quantization technique for deep learning algorithms to address these issues and introduces chaotic optimization to expedite convergence rates. It also presents a novel three-dimensional Otsu threshold segmentation method based on CNNs, which overcomes noise susceptibility in traditional two-dimensional approaches. Integrating Gray morphology and this three-dimensional Otsu threshold segmentation method results in the development of a three-dimensional Gray Otsu model. This model is the basis for designing a fitness function for the CNN algorithm, optimizing its efficacy. Experimental validation demonstrates the proposed algorithm's effectiveness, achieving an impressive 91% accuracy rate while displaying robust noise resistance and versatility. Comparative assessments against other leading AI architectures, including multilayer perceptron, radial basis function network, recurrent neural network, and long short-term memory network, affirm the superior performance achieved by the proposed approach. [ABSTRACT FROM AUTHOR]

These days, exploring information retrieval models is one of the most essential aspects of English sentence retrieval research. These models are driven by diverse retrieval mechanisms that offer varying similarity calculations and directly influence the final result ranking. However, despite decades of work due to technical constraints, deep semantic analysis has been challenging. This gap emphasizes the importance of a precise semantic understanding of information acquisition through learning approaches. Based on the above opening, this paper establishes a fast retrieval model of English sentences based on the statistical language model (SLM). First, the proposed method utilizes SLM to extract significant feature words from the corpus. These feature words are identified by analyzing co-occurrence patterns and frequency distributions within the standard. Second, it employs the N-gram model to calculate the probabilities of word occurrences based on their contextual dependencies. This framework represents feature words and their associated probabilities in a structured manner by capturing the intricate nuances of language semantics. Third, the model integrates ontology to bridge the gap between human language and machine understanding by enabling the mapping natural language expressions to conceptual entities. Finally, the suggested model retrieves English sentences through semantic matching by leveraging the comprehensive semantic framework and ontology-based search. The experimental study revealed that the proposed model demonstrated an impressive retrieval ratio of 98.5% by outperforming existing models in the comparison. Moreover, these results show that the proposed algorithm performs better than the Term Frequency-Inverse Document Frequency (TF-IDF) algorithm, and the accuracy of this algorithm is improved by 7.52% compared with TF-IDF. When the labelled corpus is very small and the unlabeled corpus is relatively large, the algorithm enhances the classifier's performance by 12.6%. This shows that the algorithm used in this paper reduces the influence of the synonym processing stage on the overall performance while retaining the advantages of high precision and accuracy of calculation results. [ABSTRACT FROM AUTHOR]

There have been a number of approaches to employ neural networks in self-adaptive systems; in many cases, generic neural networks and deep learning are utilized for this purpose. When this approach is to be applied to improve an adaptation process initially driven by logical adaptation rules, the problem is that (1) these rules represent a significant and tested body of domain knowledge, which may be lost if they are replaced by a neural network, and (2) the learning process is inherently demanding given the black-box nature and the number of weights in generic neural networks to be trained. In this paper, we introduce the rule-specific neural network method that makes it possible to transform the guard of an adaptation rule into a rule-specific neural network, the composition of which is driven by the structure of the logical predicates in the guard. Our experiments confirmed that the black box effect is eliminated, the number of weights is significantly reduced, and much faster learning is achieved whilst the accuracy is preserved. This text is an extended version of the paper presented at the ISOLA 2022 conference (Bureš et al. in Proceedings of ISOLA 2022, Rhodes, Greece, pp. 215–230, 2022). [ABSTRACT FROM AUTHOR]

Today's factories are considered as smart ecosystems with humans, machines and devices interacting with each other for efficient manufacturing of products. Industry 4.0 is a suite of enabler technologies for such smart ecosystems that allow transformation of industrial processes. When implemented, Industry 4.0 technologies have a huge impact on efficiency, productivity and profitability of businesses. The adoption and implementation of Industry 4.0, however, require to overcome a number of practical challenges, in most cases, due to the lack of modernisation and automation in place with traditional manufacturers. This paper presents a first of its kind case study for moving a traditional food manufacturer, still using the machinery more than one hundred years old, a common occurrence for small- and medium-sized businesses, to adopt the Industry 4.0 technologies. The paper reports the challenges we have encountered during the transformation process and in the development stage. The paper also presents a smart production control system that we have developed by utilising AI, machine learning, Internet of things, big data analytics, cyber-physical systems and cloud computing technologies. The system provides novel data collection, information extraction and intelligent monitoring services, enabling improved efficiency and consistency as well as reduced operational cost. The platform has been developed in real-world settings offered by an Innovate UK-funded project and has been integrated into the company's existing production facilities. In this way, the company has not been required to replace old machinery outright, but rather adapted the existing machinery to an entirely new way of operating. The proposed approach and the lessons outlined can benefit similar food manufacturing industries and other SME industries. [ABSTRACT FROM AUTHOR]

Purpose: There are millions of people who lose their life due to several types of fatal diseases. Cancer is one of the most fatal diseases which may be due to obesity, alcohol consumption, infections, ultraviolet radiation, smoking, and unhealthy lifestyles. Cancer is abnormal and uncontrolled tissue growth inside the body which may be spread to other body parts other than where it has originated. Hence it is very much required to diagnose the cancer at an early stage to provide correct and timely treatment. Also, manual diagnosis and diagnostic error may cause of the death of many patients hence much research are going on for the automatic and accurate detection of cancer at early stage. Methods: In this paper, we have done the comparative analysis of the diagnosis and recent advancement for the detection of various cancer types using traditional machine learning (ML) and deep learning (DL) models. In this study, we have included four types of cancers, brain, lung, skin, and breast and their detection using ML and DL techniques. In extensive review we have included a total of 130 pieces of literature among which 56 are of ML-based and 74 are from DL-based cancer detection techniques. Only the peer reviewed research papers published in the recent 5-year span (2018–2023) have been included for the analysis based on the parameters, year of publication, feature utilized, best model, dataset/images utilized, and best accuracy. We have reviewed ML and DL-based techniques for cancer detection separately and included accuracy as the performance evaluation metrics to maintain the homogeneity while verifying the classifier efficiency. Results: Among all the reviewed literatures, DL techniques achieved the highest accuracy of 100%, while ML techniques achieved 99.89%. The lowest accuracy achieved using DL and ML approaches were 70% and 75.48%, respectively. The difference in accuracy between the highest and lowest performing models is about 28.8% for skin cancer detection. In addition, the key findings, and challenges for each type of cancer detection using ML and DL techniques have been presented. The comparative analysis between the best performing and worst performing models, along with overall key findings and challenges, has been provided for future research purposes. Although the analysis is based on accuracy as the performance metric and various parameters, the results demonstrate a significant scope for improvement in classification efficiency. Conclusion: The paper concludes that both ML and DL techniques hold promise in the early detection of various cancer types. However, the study identifies specific challenges that need to be addressed for the widespread implementation of these techniques in clinical settings. The presented results offer valuable guidance for future research in cancer detection, emphasizing the need for continued advancements in ML and DL-based approaches to improve diagnostic accuracy and ultimately save more lives. [ABSTRACT FROM AUTHOR]

This paper explores the potential benefits of quantum coherence and quantum discord in the non-universal quantum computing model called deterministic quantum computing with one qubit (DQC1) in supervised machine learning. We show that the DQC1 model can be leveraged to develop an efficient method for estimating complex kernel functions. We demonstrate a simple relationship between coherence consumption and the kernel function, a crucial element in machine learning. The paper presents an implementation of a binary classification problem on IBM hardware using the DQC1 model and analyzes the impact of quantum coherence and hardware noise. The advantage of our proposal lies in its utilization of quantum discord, which is more resilient to noise than entanglement. [ABSTRACT FROM AUTHOR]

Music expresses thoughts and emotions in artistic form and is made up of different components including harmony, rhythm, and melody. Several musical elements are tied together during the songwriting process in order to generate melodies that are harmonic. A music melody is the essential element of any music to generate strong feelings from listeners and capture their interest. In the process of music appreciation, melody controls the emotional changes of music. It is an efficiently perceived part and the tone of a song. In recent years, Sichuan unvoiced music has developed rapidly and attracted much attention. This paper selects Sichuan unvoiced music as the main research theme and constructs a melody generation algorithm by utilizing the state-of-the-art techniques of deep learning (DL) and evolutionary algorithms (EAs) such as recurrent neural network-long short-term memory (RNN-LSTM) and genetic algorithm (GA). Firstly, this paper briefly describes the concept of DL algorithms, the deep generation model, and sequence to sequence model, as they constitute the technological foundation for this research. Secondly, this paper proposes a melody generation algorithm that utilizes RNN-LSTM for melody generation and GA for melody optimization. More specifically, the melody is generated by preprocessing data, creating, and training the RNN-LSTM model. A GA was used to determine the melodic fitness function for eight songs as the fitness function directly affects the selected termination condition. The fitness function can be thought of as either a person or an evolutionary rule. Finally, the average score of these songs, before and after evolution, is calculated, which demonstrates that the analysis and rotation creation methods are more precise and that the song's average melody score is higher. The method proposed in this study has been thoroughly compared to the existing approaches proposed in earlier studies, and it was found that the approach we propose is more effective in terms of accuracy and the average melody score. [ABSTRACT FROM AUTHOR]

Controlling complex nonlinear dynamical systems using traditional methods has always been a difficult task because the majority of systems seen in nature have intricate nonlinear mathematical relationships. Artificial neural network (ANN) models are a good option for handling such intricate nonlinear systems since they include a number of significant properties like faster learning, adaptation, parallel processing, and nonlinear mapping capabilities. Several recurrent neural networks (RNNs)-based controllers have been suggested in the literature for implementing adaptive control, but the majority of these models have extremely complex topologies and many of them are challenging to train. In this paper, an attempt is made to put forward the RNN model (called as higher-order recurrent neural network (HORNN)) which is based on a higher order Pi-Sigma neural network (PSNN) model and implemented for the indirect adaptive control of the nonlinear dynamical system. The parameters of the proposed controller are tuned using the gradient-descent-based asynchronous back-propagation (BP) method. The proposed controller consists of two additional internal feedback loop layers (denoted by F L 1 and F L 2 ) corresponding to the hidden and the output layer, respectively. The nodes present in F L 1 and F L 2 layers are having weighted connections with the hidden and the output layer neurons, respectively, and these feedback connections enrich the controller with a memory property. The second contribution of the paper is to improve the performance of the learning algorithm which is achieved by incorporating an adaptive learning rate scheme (that ensures the correct setting of the learning rate value in each iteration). Another advantage of the HORNN-based controller is that it is only provided with three inputs irrespective of the dynamics of the plant and only 3 hidden neurons are included in its hidden layer (this reduces the overall structural complexity of the proposed model). The performance of the HORNN-based controller is compared with some of the popular neural networks such as diagonal recurrent neural network (DRNN), Jordan recurrent neural network (JRNN), feed-forward neural network (FFNN), and PSNN. Through simulation experiments, it is observed that the response obtained from the plant under HORNN-based controller is found to be better as compared to responses obtained with other ANN-based controllers. Further, the instantaneous mean square error (IMSE) obtained with HORNN-based controller is quite less and is equal to 0.058 as compared to 0.077, 0.082, 1.74, 13.43, and 1.86 with DRNN, JRNN, PSNN, FFNN, and FFNN (with 30 hidden neurons)-based controllers, respectively. [ABSTRACT FROM AUTHOR]

In this paper, preliminary studies on the failure analysis of hybrid composite materials utilizing acoustic emission and machine learning are presented. The main purpose of this study was to analyze the possibilities of using machine learning techniques as a way to better cluster the data obtained from acoustic emission. In this paper, we focus on data preparation, feature extraction (Laplacian score), determination of cluster number (Caliński–Harabasz, Silhouette, and Davies–Bouldin), and testing three clustering techniques, namely K-means, fuzzy C-means, and spectral clustering. The dataset was obtained by testing fiber metal laminates—composites consisting of metal and composite layers. Two experimental tests were realized on pre-cracked rectangular specimens—one with loading in mode I and one with loading in mode II (DCB—double cantilever beam and ENF—end-notch flexural test). Elastic waves were recorded during these tests via an acoustic emission system. Preliminary studies show that the proposed method can be used successfully to cluster data obtained in this way. The obtained dataset was split into 3 clusters (for the ENF test) and 5 clusters (DCB test). In the next stages of the research campaign, based on the presented results, we intend to change the approach to semi-supervised by running additional single-cause damage tests to enhance the achieved results and enable easier damage recognition. [ABSTRACT FROM AUTHOR]

Since much emphasis has been put on the eco-friendly manufacturing process in industries, and the complex changes in market demands, this paper focuses on a green scheduling problem of automated guided vehicles (AGVs) in a flexible manufacturing system (FMS). The studied FMS consists of multi-FMCs (flexible manufacturing cells) which have many material handling needs with time constraints. Distinguished from other AGV scheduling problems in FMS, this paper concentrates on the pickup and delivery operations or even bi-handling requirements of AGVs for FMCs, ignoring the inner production process within them. To solve this problem, a bi-objective mathematical model is built trying to minimize the total tardiness and energy consumption of AGVs simultaneously. Some properties of the problem and a no-collision algorithm are developed for the potential conflicts among AGVs. Due to the NP-hard nature of the proposed problem, a hyper-heuristic (HH) algorithm based on a double deep Q network (DDQN) is introduced, which benefits from the structures of double decision networks and multi-operator. To improve the performance of the proposed algorithm, the experience pool is used to increase the convergence speed and the crowding distance, and the non-dominated sorting strategies are presented to decide the acceptance of the new generation. Besides, in the DDQN, the states and rewards of agents are designed based on the characteristics of the scheduling problem. Finally, many experiments have been conducted and the computational results reveal that the proposed DDQN-HH algorithm outperforms the other two compared algorithms in both the convergence speed and quality of solutions. [ABSTRACT FROM AUTHOR]