Search

Based on the magnetic gear effect, the field-modulated permanent-magnet machine (FMPMM) can realize the unequal pole design of the rotor PM field and the stator armature magnetic field. With the advantages of high torque density and high efficiency, the FMPMM has been widely studied in low-speed direct-drive applications. As a kind of machine excited by PMs, the performance of the FMPMM was affected by the demagnetization state. However, the method for establishing the FMPMM demagnetization model based on a finite element analysis (FEA) presented some problems, including tedious repeated modeling work and long calculation time-consuming under fine subdivision. Therefore, in this paper, a six-phase surface-mounted FMPMM was taken as an example, and an equivalent magnetic network (EMN) model was proposed for evaluating the machine performance under demagnetization. In order to realize the rapid establishing EMN models under diverse demagnetization types, the variable coercivity of PM was introduced. Furthermore, for the purpose of improving the calculation accuracy and shortening the calculation time, the least square method was used in fitting and analyzing the discrete results. Then, in order to verify the validity of the proposed EMN model, a prototype was fabricated and a testing platform was built. The air-gap flux density and the no-load back EMF obtained by the FEA, the proposed EMN model, and the experimental testing were compared. The results showed that the proposed EMN model can realize the rapid modeling and accurate analysis of the six-phase surface-mounted FMPMM under diverse demagnetization types. [ABSTRACT FROM AUTHOR]

Milling force is one of the key output variables related to milling efficiency and accuracy. However, the time-consuming identification of milling forces cannot be evaluated online for machining performance during the milling process. In this paper, a novel millisecond-level method is developed to identify the milling forces online using acceleration signals. Firstly, the milling force identification problem is transformed into an ill-posed problem in the time-domain convolution framework. Next, the advantages of the regularization method for the ill-posed problem are analyzed. Then, the properties of the transfer matrix and the regularization operator are used to reduce the milling force identification time. Lastly, the proposed method is verified with several sets of milling experiments, and the results show that the accuracy of milling force identification is acceptable and the identification time is less than 10 ms. [ABSTRACT FROM AUTHOR]

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

Cloud computing has become a leading technology for IT infrastructure, with many companies migrating their services to cloud servers in recent years. As cloud services continue to expand, the issue of cloud monitoring has become increasingly important. One important metric to monitor is CPU steal time, which measures the amount of time a virtual CPU waits for the actual CPU. In this study, we focus on the impact of CPU steal time on virtual machine performance and the potential problems that can arise. We implement our work using an OpenStack-based cloud environment and investigate intrusive and non-intrusive monitoring methods. Our analysis provides insights into the importance of CPU steal time monitoring and its impact on cloud performance. [ABSTRACT FROM AUTHOR]

Lithium-ion batteries are widely used in electric vehicles, smartphones, and energy storage devices due to their high power and light weight. The goal of this study is to predict the remaining capacity of a lithium-ion battery and evaluate its performance through three machine learning models: linear regression, decision tree, and random forest, and two deep learning models: neural network and ensemble model. Mean squared error (MSE), mean absolute error (MAE), coefficient of determination (R-squared), and root mean squared error (RMSE) were used to measure prediction accuracy. For the evaluation of the artificial intelligence model, the dataset was downloaded and integrated with measurement data of the CS2 lithium-ion battery provided by the University of Maryland College of Engineering. As a result of the study, the RMSE of the linear regression model was 0.045, the decision tree model was 0.038, the random forest model was 0.034, the neural network model was 0.032, and the ensemble model was 0.030. According to the measured values, the ensemble model showed the best predictive performance, followed by the neural network model. Decision tree and random forest models also showed very good performance, and the linear regression model showed relatively poor predictive performance compared to the other models. [ABSTRACT FROM AUTHOR]

Sentiment Analysis (SA) using machine learning approaches is a fascinating field of research to unravel the opinion expressed by online users. It deals with the classification problem of opinions expressed on a product or service in review texts as positive or negative or neutral. Classification beyond binary classes poses even greater challenges using machine learning. In this paper two sets of approaches, first set of approach on traditional machine learning and second one on deep learning technique was and evaluated and analyzed for multiclass polarity classification at fine-grain level on cross-domain review dataset. The analysis of performance of traditional machine learning approaches that comprise ensemble models and incorporate semantic and diverse statistical features, shows that ensemble models are better in their category. With the proposed sentiment-document model coupled with the robustness of the proposed ensemble models, we were able to investigate and establish that the document polarity could serve as the latent mutual information that could leverage the model performance of predicting sentiments at sentence and document levels. Furthermore, the evaluation of deep learning models infers that they are more viable option for multi-class polarity classification of cross-domain dataset. They outperformed the traditional approaches based on feature extraction and feature engineering and achieved better accuracy and F1 scores. [ABSTRACT FROM AUTHOR]

Purpose: With the in-depth application of machine learning(ML) in clinical practice, it has been used to predict the mortality risk in patients with traumatic brain injuries(TBI). However, there are disputes over its predictive accuracy. Therefore, we implemented this systematic review and meta-analysis, to explore the predictive value of ML for TBI. Methodology: We systematically retrieved literature published in PubMed, Embase.com, Cochrane, and Web of Science as of November 27, 2022. The prediction model risk of bias(ROB) assessment tool (PROBAST) was used to assess the ROB of models and the applicability of reviewed questions. The random-effects model was adopted for the meta-analysis of the C-index and accuracy of ML models, and a bivariate mixed-effects model for the meta-analysis of the sensitivity and specificity. Result: A total of 47 papers were eligible, including 156 model, with 122 newly developed ML models and 34 clinically recommended mature tools. There were 98 ML models predicting the in-hospital mortality in patients with TBI; the pooled C-index, sensitivity, and specificity were 0.86 (95% CI: 0.84, 0.87), 0.79 (95% CI: 0.75, 0.82), and 0.89 (95% CI: 0.86, 0.92), respectively. There were 24 ML models predicting the out-of-hospital mortality; the pooled C-index, sensitivity, and specificity were 0.83 (95% CI: 0.81, 0.85), 0.74 (95% CI: 0.67, 0.81), and 0.75 (95% CI: 0.66, 0.82), respectively. According to multivariate analysis, GCS score, age, CT classification, pupil size/light reflex, glucose, and systolic blood pressure (SBP) exerted the greatest impact on the model performance. Conclusion: According to the systematic review and meta-analysis, ML models are relatively accurate in predicting the mortality of TBI. A single model often outperforms traditional scoring tools, but the pooled accuracy of models is close to that of traditional scoring tools. The key factors related to model performance include the accepted clinical variables of TBI and the use of CT imaging. [ABSTRACT FROM AUTHOR]

The high-precision machine tool's dynamic, static, and rigid nature directly affects the machining efficiency and surface quality. Static and dynamic analyses are essential for the design and improvement of precision machine to ensure good tool performance under difficult and demanding machining conditions. In this study, the performance of a high-precision machine tool was analyzed using its virtual model created using CAD. Static and model analysis using ANSYS Workbench software was conducted to establish the tool's static deformation and static stiffness. Furthermore, the static and dynamic characteristics of the tool were explored using a finite element modeling approach to study their performance. In particular, the structure, static force, modal, frequency spectrum, and topology optimization of machine tools were primarily analyzed. Using model analysis, we found the first four different frequencies (22.5, 28.9, 40.6, and 47.4 Hz) and vibration type, which suggested of a weak link. Further static structural analysis revealed that the deformation of the spindle was 67.26 μm. An experimental static rigidity analysis was performed, and the experimental deformation values of the tool and spindle were obtained. The static and dynamic characteristics, as well as the accuracy and efficiency of the finite element model, were verified by comparing the data with the finite element analysis (FEA) results. Subsequently, we modified the settings and analysis model to ensure that the analysis results were consistent with the experimental findings. The error between the two results was within 1.56%. For an applied load of 5000 N on the spindle nose, the tool nose transient response was 0.5 s based on transient analysis. Under the condition that the structural deformation is as constant as possible, the lightweight structure may achieve the minimum weight and enhance the natural frequency; thus, the ideal structure will be obtained, and finite element analysis will then be performed. The optimal conditions for topology optimization include a lightweight structure, reduced structural deformation, and increased natural frequency of the structure. The developed method improves structural optimization, increases the ability of the product to be manufactured, and offers designers a variety of price-effective options. [ABSTRACT FROM AUTHOR]

Demonstrating power performance in real conditions is vital for verifying tidal turbine design. The International Electrotechnical Commission's Technical Specification IEC TS 62600-200 provides guidance for tidal developers to evaluate machine performance. This study evaluates the performance of the operational 1 MW DEEP-Gen IV commercial-scale tidal turbine deployed during the ReDAPT project at the European Marine Energy Centre's tidal test site in Orkney, Scotland. IEC TS 62600-200 states that the power performance should be measured relative to two independently located current profilers deployed in one of two orientations; 'in-line' (preferred) or 'adjacent' (least preferred), relative to the turbine. Two measurement campaigns are used to assess the impact of instrument placement on the measured power curve and annual energy production (AEP). The ambient current flow and vertical profile form differed at each current profiler location (due to topography and bathymetry), leading to AEP variability. The results reinforce the IEC TS preferred orientation for instrument placement, with inline measurements showing smaller variation in AEP estimates compared with adjacent measurements. An additional analysis assesses the sensitivity of AEP estimates on the vertical alignment of the flow profiles with rotor positioning. For this we consider three turbine scenarios: the DEEP-Gen IV and two modelled turbine rotors occupying different regions in the water column. Results show that for regions of high vertical shear, AEP estimates can be misrepresented by up to 5.5% under an imposed vertical misalignment of 2 m. This work has quantified, at this site, the variation in AEP estimates from different flow measurement campaigns informed by IEC TS 62600-200 and has shown the need to ensure good vertical alignment between measurements and turbine rotor positioning. • The measured power curve for TEC is sensitive to the flow reference used. • ADCPs placed in-line to the TEC provide a better reference than adjacently placed. • In-line flow measurements varied despite being placed at IEC prescribed distance. • Large flow structures impacted measurements 30 m from the TEC location. • Vertical ADCP misalignment (±2 m) to TEC swept area can impact AEP by up to 5.5%. [ABSTRACT FROM AUTHOR]

In order to predict the thermal deformation of CNC machine tool spindle system more accurately, a method based on multi-source heterogeneous information fusion is proposed. Aiming at the problem that it is difficult to obtain the global information of thermal deformation with a single type of information source, the effective prediction of thermal deformation of spindle system in machining center is realized under the fusion of temperature and vibration signals. First, the combined denoising method of empirical mode decomposition (EMD) and wavelet threshold is used to preprocess the vibration data. Then, the time-domain, frequency-domain, and time-frequency features of vibration signals are extracted, and the feature dimension is reduced based on correlation analysis and kernel principal component analysis (KPCA). After dimensionality reduction, the vibration data and temperature information are fused in the eigenmatrix. The nonlinear prediction of thermal deformation is studied by support vector regression for grid search parameter optimization (GS-SVR) and support vector regression for particle swarm optimization (PSO-SVR) methods. In order to realize the information acquisition of multi-channel temperature, vibration, and verify the effectiveness of the prediction model in the case of information fusion, a specific machining center is taken as the research object and experiments are performed based on multi-channel and heterogeneous signal acquisition. Finally, the prediction results based on different information sources are compared and analyzed. The results show that the thermal deformation of the machine tool obtained by the multi-source heterogeneous information fusion method is consistent with the actual test results. As compared with the predicted performance using only temperature information, the mean square error (MSE) decreased by 0.1663 µm. Therefore, the temperature - vibration information fusion model has higher accuracy in terms of predicting the thermal deformation of the spindle system model. [ABSTRACT FROM AUTHOR]

The fault tolerance proprieties of multiphase induction machines make them candidate technologies for future electrified transport systems. Indeed, the use of multiphase drives is primarily recommended for their redundancy, which allows them to handle increasing power demands as well as provide fault-tolerant operation. One of the most common failures in polyphase induction machines is open-phase fault (OPF). Under this degraded mode, machine performance deteriorates with increased torque ripples and copper losses. This work investigates the fault-tolerant operation of dual star induction machines (DSIM) with two connected neutral points in which a new post-fault operation technique is introduced to manage the OPFs. The machine was modeled on the natural reference frame, and no transformations were used. The phase opening is caused by inserting a high-value resistor in series with the faulty phase, which cancels the current flowing in the latter. A post-fault reconfiguration strategy that reorganizes the power converter in order to cancel torque ripples by exploiting the principle of multi-biphase modeling of the DSIM is proposed. The performance of the proposed reconfiguration strategy was verified through detailed simulation results. [ABSTRACT FROM AUTHOR]

The Internet of Things is rapidly growing with the demand for low-power, long-range wireless communication technologies. Long Range Wide Area Network (LoRaWAN) is one such technology that has gained significant attention in recent years due to its ability to provide long-range communication with low power consumption. One of the main issues in LoRaWAN is the efficient utilization of radio resources (e.g., spreading factor and transmission power) by the end devices. To solve the resource allocation issue, machine learning (ML) methods have been used to improve the LoRaWAN network performance. The primary aim of this survey paper is to study and examine the issue of resource management in LoRaWAN that has been resolved through state-of-the-art ML methods. Further, this survey presents the publicly available LoRaWAN frameworks that could be utilized for dataset collection, discusses the required features for efficient resource management with suggested ML methods, and highlights the existing publicly available datasets. The survey also explores and evaluates the Network Simulator-3-based ML frameworks that can be leveraged for efficient resource management. Finally, future recommendations regarding the applicability of the ML applications for resource management in LoRaWAN are illustrated, providing a comprehensive guide for researchers and practitioners interested in applying ML to improve the performance of the LoRaWAN network. [ABSTRACT FROM AUTHOR]

Algorithms for machine learning have found extensive use in numerous fields and applications. One important aspect of effectively utilizing these algorithms is tuning the hyperparameters to match the specific task at hand. The selection and configuration of hyperparameters directly impact the performance of machine learning models. Achieving optimal hyperparameter settings often requires a deep understanding of the underlying models and the appropriate optimization techniques. While there are many automatic optimization techniques available, each with its own advantages and disadvantages, this article focuses on hyperparameter optimization for well-known machine learning models. It explores cutting-edge optimization methods such as metaheuristic algorithms, deep learning-based optimization, Bayesian optimization, and quantum optimization, and our paper focused mainly on metaheuristic and Bayesian optimization techniques and provides guidance on applying them to different machine learning algorithms. The article also presents real-world applications of hyperparameter optimization by conducting tests on spatial data collections for landslide susceptibility mapping. Based on the experiment's results, both Bayesian optimization and metaheuristic algorithms showed promising performance compared to baseline algorithms. For instance, the metaheuristic algorithm boosted the random forest model's overall accuracy by 5% and 3%, respectively, from baseline optimization methods GS and RS, and by 4% and 2% from baseline optimization methods GA and PSO. Additionally, for models like KNN and SVM, Bayesian methods with Gaussian processes had good results. When compared to the baseline algorithms RS and GS, the accuracy of the KNN model was enhanced by BO-TPE by 1% and 11%, respectively, and by BO-GP by 2% and 12%, respectively. For SVM, BO-TPE outperformed GS and RS by 6% in terms of performance, while BO-GP improved results by 5%. The paper thoroughly discusses the reasons behind the efficiency of these algorithms. By successfully identifying appropriate hyperparameter configurations, this research paper aims to assist researchers, spatial data analysts, and industrial users in developing machine learning models more effectively. The findings and insights provided in this paper can contribute to enhancing the performance and applicability of machine learning algorithms in various domains. [ABSTRACT FROM AUTHOR]

In this work, we address the problem of improving the classification performance of machine learning models, especially in the presence of noisy and outlier data. To this end, we first innovatively design a generalized adaptive robust loss function called  V θ (x) . Intuitively,  V θ (x)  can improve the robustness of the model by selecting different robust loss functions for different learning tasks during the learning process via the adaptive parameter  θ. Compared with other robust loss functions,  V θ (x)  has some desirable salient properties, such as symmetry, boundedness, robustness, nonconvexity, and adaptivity, making it suitable for a wide range of machine learning applications. Secondly, a new robust semi-supervised learning framework for pattern classification is proposed. In this learning framework, the proposed robust loss function  V θ (x)  and capped  L 2 , p -norm robust distance metric are introduced to improve the robustness and generalization performance of the model, especially when the outliers are far from the normal data distributions. Simultaneously, based on this learning framework, the Welsch manifold robust twin bounded support vector machine (WMRTBSVM) and its least-squares version are developed. Finally, two effective iterative optimization algorithms are designed, their convergence is proved, and their complexity is calculated. Experimental results on several datasets with different noise settings and different evaluation criteria show that our methods have better classification performance and robustness. With the Cancer dataset, when there is no noise, the classification accuracy of our proposed methods is  94.17 %  and  95.62 % , respectively. When the Gaussian noise is  50 % , the classification accuracy of our proposed methods is  91.76 %  and  90.59 % , respectively, demonstrating that our method has satisfactory classification performance and robustness. [ABSTRACT FROM AUTHOR]

Distribution System Operators (DSOs) and Aggregators benefit from novel Energy Generation Forecasting (EGF) approaches. Improved forecasting accuracy may make it easier to deal with energy imbalances between production and consumption. It also aids operations such as Demand Response (DR) management in Smart Grid architecture. This work aims to develop and test a new solution for EGF. It combines various methodologies running EGF tests on historical data from buildings. The experimentation yields different data resolutions (15 min, one hour, one day, etc.) while reporting accuracy errors. The optimal forecasting technique should be relevant to a variety of forecasting applications in a trial-and-error manner, while utilizing different forecasting strategies, ensemble approaches, and algorithms. The final forecasting evaluation incorporates performance metrics such as coefficient of determination ( R 2 ), Mean Absolute Error (MAE), Mean Squared Error (MSE) and Root Mean Squared Error (RMSE), presenting a comparative analysis of results. [ABSTRACT FROM AUTHOR]

Twin-field quantum key distribution (TF-QKD) has the advantage of beating the rate-loss limit (PLOB bound) for a repeaterless quantum key distribution (QKD) system. In practice, parameter optimization is of great significance in maximizing the secret key rate. Nevertheless, traditional local search algorithms (LSA) are often time-consuming and limited by the computing capabilities of devices. In this paper, we use the machine learning method instead of LSA to directly predict the optimal parameters for TF-QKD system. Specifically, three neural networks, namely back propagation neural network, radial basis function neural network, and generalized regression neural network, are trained and evaluated. The performance of neural networks and LSA in optimizing parameters is discussed and analyzed in this study. It is proved that the performance of machine learning-based prediction method is comparable to LSA, but the calculation time is shortened by 6 orders of magnitude. Furthermore, a comprehensive comparison of three networks in terms of prediction accuracy and time consumption is conducted, serving as a guide for selecting the most suitable network to optimize parameters in a practical TF-QKD system with different optimization requirements. [ABSTRACT FROM AUTHOR]

Introduction: There is some debate on the relationship between essential tremor with rest tremor (rET) and the classic ET syndrome, and only few MRI studies compared ET and rET patients. This study aimed to explore structural cortical differences between ET and rET, to improve the knowledge of these tremor syndromes. Methods: Thirty-three ET patients, 30 rET patients and 45 control subjects (HC) were enrolled. Several MR morphometric variables (thickness, surface area, volume, roughness, mean curvature) of brain cortical regions were extracted using Freesurfer on T1-weighted images and compared among groups. The performance of a machine learning approach (XGBoost) using the extracted morphometric features was tested in discriminating between ET and rET patients. Results: rET patients showed increased roughness and mean curvature in some fronto-temporal areas compared with HC and ET, and these metrics significantly correlated with cognitive scores. Cortical volume in the left pars opercularis was also lower in rET than in ET patients. No differences were found between ET and HC. XGBoost discriminated between rET and ET with mean AUC of 0.86 ± 0.11 in cross-validation analysis, using a model based on cortical volume. Cortical volume in the left pars opercularis was the most informative feature for classification between the two ET groups. Conclusion: Our study demonstrated higher cortical involvement in fronto-temporal areas in rET than in ET patients, which may be linked to the cognitive status. A machine learning approach based on MR volumetric data demonstrated that these two ET subtypes can be distinguished using structural cortical features. [ABSTRACT FROM AUTHOR]

This study aimed to explore the effectiveness of predicting disease activity in patients with inflammatory bowel disease (IBD), using machine learning (ML) models. A retrospective research was undertaken on IBD patients who were admitted into the First Affiliated Hospital of Wenzhou Medical University between September 2011 and September 2019. At first, data were randomly split into a 3:1 ratio of training to test set. The least absolute shrinkage and selection operator (LASSO) algorithm was applied to reduce the dimension of variables. These variables were used to generate seven ML algorithms, namely random forests (RFs), adaptive boosting (AdaBoost), K-nearest neighbors (KNNs), support vector machines (SVMs), naïve Bayes (NB), ridge regression, and eXtreme gradient boosting (XGBoost) to train to predict disease activity in IBD patients. SHapley Additive exPlanation (SHAP) analysis was performed to rank variable importance. A total of 876 participants with IBD, consisting of 275 ulcerative colitis (UC) and 601 Crohn's disease (CD), were retrospectively enrolled in the study. Thirty-three variables were obtained from the clinical characteristics and laboratory tests of the participants. Finally, after LASSO analysis, 11 and 5 variables were screened out to construct ML models for CD and UC, respectively. All seven ML models performed well in predicting disease activity in the CD and UC test sets. Among these ML models, SVM was more effective in predicting disease activity in the CD group, whose AUC reached 0.975, sensitivity 0.947, specificity 0.920, and accuracy 0.933. AdaBoost performed best for the UC group, with an AUC of 0.911, sensitivity 0.844, specificity 0.875, and accuracy 0.855. ML algorithms were available and capable of predicting disease activity in IBD patients. Based on clinical and laboratory variables, ML algorithms demonstrate great promise in guiding physicians' decision-making. [ABSTRACT FROM AUTHOR]

Ultrasonic vibration–assisted electrical discharge machining (UEDM) was studied to improve the machining performance for TC4 micro-holes. First, a computational fluid dynamic (CFD) model of inter-electrode gap for the micro-hole was established using COMSOL Multiphysics, and the effect of ultrasonic amplitude on the distribution of fluid and debris in the inter-electrode gap was analyzed. The experimental study of UEDM for machining performance of TC4 micro-hole under different machining parameters was conducted by using the controlled variable method. Results showed that the ultrasonic vibration of the electrode is beneficial to improve the machining performance for micro-hole. Under the optimal processing parameters, the material removal rate is improved four times, and RTWR and θ are reduced by about 50%. Finally, the neural network and genetic algorithm were used to predict and optimize the experimental results. The average error is 4.1% between the optimal processing parameters [Xn] obtained by the above method and from the experiment. [ABSTRACT FROM AUTHOR]

Nickel-based superalloys (Ni-alloys) are widely used in flight critical aeroengine components because of their excellent material properties at high temperatures such as yield strength, ductility, and creep resistance. However, these desirable high-temperature properties also make Ni-alloys very difficult to machine. This paper provides an overview and benchmarking of various constitutive models to provide the process modeling community with an objective comparison between various calibrated material models to increase the accuracy of process model predictions for machining of Ni-alloys. Various studies involving the Johnson-Cook model and the calibration of its constants in finite element simulations are discussed. It was found that significant discrepancies exist between researchers' approaches to calibrating constitutive models. To this end, various "physics-based" models are discussed as an alternative to widely used "phenomenological" models like the Johnson-Cook model, supplemented by a discussion on the more precise inverse method for constitutive model calibration. This paper also provides a comprehensive overview of pedigreed physical material properties for a range of Ni-alloys-the variation of thermal properties and thermally induced stresses over machining temperature regimes are modeled for a variety of Ni-alloys. The chemical compositions and applications for a range of relevant Ni-alloys are also explored. Overall, this paper identifies the need for more comprehensive analysis and process-specific (e.g., in-situ) characterization of thermomechanical properties for difficult-to-machine Ni-alloys to improve machining performance and aeroengine component quality. [ABSTRACT FROM AUTHOR]

In photolithographic processes, nanometer-level-precision wavefront-aberration models enable the machine to be able to meet the overlay (OVL) drift and critical dimension (CD) specifications. Software control algorithms take as input these models and correct any expected wavefront imperfections before reaching the wafer. In such way, a near-optimal image is exposed on the wafer surface. Optimizing the parameters of these models, however, involves several time costly sensor measurements which reduce the throughput performance of the machine in terms of exposed wafers per hour. In that case, photolithography machines come across the trade-off between throughput and quality. Therefore, one of the most common optimal experimental design (OED) problems in photolithography machines (and not only) is how to choose the minimum amount of sensor measurements that will provide the maximum amount of information. Additionally, each sensor measurement corresponds to a point on the wafer surface and therefore we must measure uniformly around the wafer surface as well. In order to solve this problem, we propose a sensor mark selection algorithm which exploits genetic algorithms. The proposed solution first selects a pool of points that qualify as candidates to be selected in order to meet the uniformity constraint. Then, the point that provides the maximum amount of information, quantified by the Fisher-based criteria of G-, D-, and A-optimality, is selected and added to the measurement scheme. This process, however, is considered "greedy", and for this reason, genetic algorithms (GA) are exploited to further improve the solution. By repeating in parallel the "greedy" part several times, we obtain an initial population that will be the input to our GA. This meta-heuristic approach outperforms the "greedy" approach significantly. The proposed solution is applied in a real life semiconductors industry use case and achieves interesting industry as well as academical results. [ABSTRACT FROM AUTHOR]

Recently, several indirectly controlled sliding robots have been designed to achieve efficient and stable locomotion on slippery surfaces, and numerical simulations proved their possibility. However, it is difficult to achieve the same performance on a real machine because the wobbling mass, composed of springs and dampers, shows unexpected behavior when it is moved in translation. In this study, we propose a model of an arc-shaped sliding locomotion robot with a rotating wobbling mass. Specifically, the position of the center of gravity of the robot was changed, and the pull-in phenomenon due to the rotational motion of the wobbling mass is utilized. By rotating the wobbling mass, the sliding motion is realized while maintaining a strong propulsive force. First, we performed experimental validation of the proposed new mechanism. Subsequently, a detailed mathematical model was constructed for numerical analysis. Finally, the motion performance was optimized by the Bayesian optimization method. [ABSTRACT FROM AUTHOR]

Improving the predictive performance of machine learning models is the desired goal in many tasks and domains. The predictive performance of the learning algorithm is directly affected by the input features it receives. Feature augmentation is aimed at enhancing the quality of models by adding informative features to the original data. Explainable AI methods are typically used to explain the results of machine learning models. Recently, these methods have also been used to improve models' predictive performance. In this study, we examine the benefit of incorporating the explanations obtained by an explainable AI method as augmented features. In particular, we propose SFA — Shapley-Based feature augmentation, a two-stage ensemble learning method that uses out-of-fold predictions and their corresponding Shapley values as augmented features for each instance. Shapley values, which are obtained without domain expertise, reflect the importance of the original features to each prediction and consider their interactions with all other features. Experimental results demonstrate the superiority of our proposed method, SFA, against several feature augmentation methods on multiple public datasets with various characteristics. [Display omitted] • Improving the performance of decision forests is considered a difficult task. • Adding informative features can enhance the quality of models. • Incorporation of explanations obtained using XAI methods as augmented features. • SFA — a two-stage ensemble method that uses OOF predictions and Shapley values. • Experiments demonstrate improvements in predictive performance using SFA. [ABSTRACT FROM AUTHOR]

Anomalies in electricity consumption can have a significant impact on the efficiency and dependability of power systems. Therefore, precise and prompt anomaly detection is essential for ensuring efficient system operation. This research paper presents an in-depth analysis of various machine-learning approaches for detecting abnormalities in Indian electricity consumption data. Specifically, the performance of three widely used machine learning algorithms, namely eXtreme Gradient Boosting, (XGBoost) Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU), is evaluated. The efficacy of each technique is assessed using performance indicators, including mean absolute error (MAE), mean squared error (MSE), and coefficient of determination (R2). The findings of this study provide valuable insights into the most effective machine-learning methods for identifying anomalies in Indian energy usage data. Moreover, this research contributes to the advancement of power grid anomaly detection technologies, leading to more effective systems. The anticipated outcome of this effort is an increase in the dependability and efficiency of the electrical supply system, benefiting both power system operators and end users. [ABSTRACT FROM AUTHOR]

Polyamide PA66 has been adopted by a variety of industries, and engineering fields. It is used in machinery part production due to its good properties. Machining is the most commonly used processing technique when high quality of part dimension and surface is required. There is a lack of knowledge about the impact of the tool’s rake angles when turning polyamide PA66, therefore, this study aims to define an optimal condition that can provide the highest performance in machining polyamide PA66 at the lowest cutting force. The impact of the tool’s side rake angle, back rake angle, and cutting depth on cutting force was studied during turning polyamide PA66 with the HSS tool. Three levels were considered for each variable and Taguchi's Orthogonal Array (OA) was used to design nine test configurations. The tests were performed experimentally on a conventional lathe machine. The resultant cutting force was calculated as the response data. The values were converted to signal-to-noise (S/N) ratio to facilitate the analysis using the Taguchi method and analysis of variance (ANOVA). Accordingly, the cutting depth showed the greatest impact on cutting force (57.12%), followed by the side rake angle (27.9%) and back rake angle (8.21%). An optimal condition set to turn polyamide PA66 at the lowest cutting force (Fc) is identified as 1 mm depth of cut, side rake angle αs = 21°, and back-rake angle αb = 8°. Finally, the optimal condition set was evaluated by conformation tests, and the results agreed with the calculations to a large extent. [ABSTRACT FROM AUTHOR]

During the operation of the guide rail pair, the accuracy of the guide rail pair is decreased by friction and wear, and the overall performance of the machine tool is directly affected. The influence of surface micro-features of the slider raceway on the dynamic friction factor of the guide rail pair is analyzed in this paper. The relationship between the geometric characteristics of the surface micro-features and the dynamic friction factor is studied, and a prediction model of the dynamic friction factor is established. Meanwhile, the correctness and rationality of the predicted model is verified by the orthogonal experiments of the friction force. It is found that the dynamic friction factor increases with the increase of the groove depth, the height of the peaks and peak–valleys on the surface of the slider raceway. The above research provides a basis for the study of the friction and wear characteristics of guide rail pairs and lays a foundation for improving its friction and wear performance. [ABSTRACT FROM AUTHOR]

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

Hyperparameter optimization (HPO) is a necessary step to ensure the best possible performance of Machine Learning (ML) algorithms. Several methods have been developed to perform HPO; most of these are focused on optimizing one performance measure (usually an error-based measure), and the literature on such single-objective HPO problems is vast. Recently, though, algorithms have appeared that focus on optimizing multiple conflicting objectives simultaneously. This article presents a systematic survey of the literature published between 2014 and 2020 on multi-objective HPO algorithms, distinguishing between metaheuristic-based algorithms, metamodel-based algorithms and approaches using a mixture of both. We also discuss the quality metrics used to compare multi-objective HPO procedures and present future research directions. [ABSTRACT FROM AUTHOR]

The Kainji hydropower station is a seven turbo-alternator station that for many years served as the base load supply for the Nigerian power grid. Over 200,000 pieces of data about the performance of the machines were used to estimate values of the failure and repair rates for each machine and a Markov steady-state model of the plant was constructed to determine the probability output of the turbines. This result showed that Kaplan turbine (KT) 12 is prone to failure compared to any other KT unit in the hydropower plant. Also, the clusters of probability that define the system state due to the different output capacities of the units show that the hydropower plant has not performed to its maximum capacity, further evaluation shows that 60% of the KT machine units are operating which is consistent with the observed robustness of the output. The model not only conforms to observations but reasonably provide a means of studying the effects of different actions that may be taken to improve the performance of hydropower plant. [ABSTRACT FROM AUTHOR]

Purpose: Quality control (QC) is performed routinely through professional guidelines. However, the recommended QC frequency may not be optimal among different institutional settings. Here we propose a novel method for determining the optimal QC frequency using risk matrix (RM) analysis. Methods and materials: A newly installed Magnetic Resonance linac (MR‐linac) was chosen as the testing platform and six routine QC items were investigated. Failures of these QC items can adversely affect treatment outcome for the patient. Accordingly, each QC item with its assigned frequency forms a unique failure mode (FM). Using FM‐effect analysis (FMEA), the severity (S), occurrence (O), and detection (D) of each FM was obtained. Next, S and D based on RM was used to determine the appropriate QC frequency. Finally, the performance of new frequency for each QC item was evaluated using the metric E = O/D. Results: One new QC frequency was the same as the old frequency, two new QC frequencies were less than the old ones, and three new QC frequencies were higher than the old ones. For six QC items, E values at the new frequencies were not less than their values at the old frequencies. This indicates that the risk of machine failure is reduced at the new QC frequencies. Conclusions: The application of RM analysis provides a useful tool for determining the optimal frequencies for routine linac QC. This study demonstrated that linac QC can be performed in a way that maintains high performance of the treatment machine in a radiotherapy clinic. [ABSTRACT FROM AUTHOR]

Background: Accurately predicting the risk of atherosclerotic cardiovascular disease (ASCVD) is crucial for implementing individualized prevention strategies and improving patient outcomes. Our objective is to develop machine learning (ML)-based models for predicting ASCVD risk in a prospective Chinese population and compare their performance with conventional regression models. Methods: A hybrid dataset consisting of 551 features was used, including 98 demographic, behavioral, and psychological features, 444 Electrocardiograph (ECG) features, and 9 Echocardiography (Echo) features. Seven machine learning (ML)-based models were trained, validated, and tested after selecting the 30 most informative features. We compared the discrimination, calibration, net benefit, and net reclassification improvement (NRI) of the ML models with those of conventional ASCVD risk calculators, such as the Pooled Cohort Equations (PCE) and Prediction for ASCVD Risk in China (China-PAR). Results: The study included 9,609 participants (mean age 53.4 ± 10.4 years, 53.7% female), and during a median follow-up of 4.7 years, 431 (4.5%) participants developed ASCVD. In the testing set, the final ML-based ANN model outperformed PCE, China-PAR, recalibrated PCE, and recalibrated China-PAR in predicting ASCVD. This was demonstrated by the model's higher area under the curve (AUC) of 0.800, compared to 0.777, 0.780, 0.779, and 0.779 for the other models, respectively. Additionally, the model had a lower Hosmer–Lemeshow χ2 of 9.1, compared to 37.3, 67.6, 126.6, and 18.6 for the other models. The net benefit at a threshold of 5% was also higher for the ML-based ANN model at 0.017, compared to 0.016, 0.013, 0.017, and 0.016 for the other models, respectively. Furthermore, the NRI was 0.089 for the ML-based ANN model, while it was 0.355, 0.098, and 0.088 for PCE, China-PAR, and recalibrated PCE, respectively. Conclusions: Compared to conventional regression ASCVD risk calculators, such as PCE and China-PAR, the ANN prediction model may help optimize identification of individuals at heightened cardiovascular risk by flexibly incorporating a wider range of potential predictors. The findings may help guide clinical decision-making and ultimately contribute to ASCVD prevention and management. [ABSTRACT FROM AUTHOR]

Background: Measles has high primary reproductive number, extremely infectious and ranked second to malaria in terms of disease burden in Ghana. Owing to the disease's high infectious rate, making early diagnosis based on an accurate system can help limit the spread of the disease. Studies have been conducted to derive models to serve as preliminary tools for early detection. However, these derived models are based on traditional methods, which may be limited in terms of model sensitivity and prediction power. This study focuses on comparing the performance of five machine learning classification techniques with a traditional method for predicting measles patients in Ghana. The study was an analytical cross-sectional design of suspected measles cases in Ghana. Results: The performance of six classifiers were compared and the random forest (RF) model demonstrated better performance among other models. The RF model achieved the highest sensitivity (0.88) specificity (0.96), ROC (0.92) and total accuracy (0.92). Conclusions: Our findings showed that, despite all the six methods had good performance in classifying measles patients, the RF model outperformed all the other classifiers in terms of different criteria in prediction accuracy. Accordingly, this approach is an effective classifier for predicting measles in the early stage. [ABSTRACT FROM AUTHOR]

Background: Machine learning (ML) methods to build prediction models starting from electrocardiogram (ECG) signals are an emerging research field. The aim of the present study is to investigate the performances of two ML approaches based on ECGs for the prediction of new-onset atrial fibrillation (AF), in terms of discrimination, calibration and sample size dependence. Methods: We trained two models to predict new-onset AF: a convolutional neural network (CNN), that takes as input the raw ECG signals, and an eXtreme Gradient Boosting model (XGB), that uses the signal's extracted features. A penalized logistic regression model (LR) was used as a benchmark. Discrimination was evaluated with the area under the ROC curve, while calibration with the integrated calibration index. We investigated the dependence of models' performances on the sample size and on class imbalance corrections introduced with random under-sampling. Results: CNN's discrimination was the most affected by the sample size, outperforming XGB and LR only around n = 10.000 observations. Calibration showed only a small dependence on the sample size for all the models considered. Balancing the training set with random undersampling did not improve discrimination in any of the models. Instead, the main effect of imbalance corrections was to worsen the models' calibration (for CNN, integrated calibration index from 0.014 [0.01, 0.018] to 0.17 [0.16, 0.19]). The sample size emerged as a fundamental point for developing the CNN model, especially in terms of discrimination (AUC = 0.75 [0.73, 0.77] when n = 10.000, AUC = 0.80 [0.79, 0.81] when n = 150.000). The effect of the sample size on the other two models was weaker. Imbalance corrections led to poorly calibrated models, for all the approaches considered, reducing the clinical utility of the models. Conclusions: Our results suggest that the choice of approach in the analysis of ECG should be based on the amount of data available, preferring more standard models for small datasets. Moreover, imbalance correction methods should be avoided when developing clinical prediction models, where calibration is crucial. [ABSTRACT FROM AUTHOR]

Phasor Measurement Units (PMU) are capable to generate multi-dimensional time series data, which is one of the most important parts for monitoring power system operation. However, various internal and external factors frequently cause the system to generate anomalous data randomly, so we expect to clean and re-fill the raw PMU data with anomalies to provide support for further advanced applications such as situational awareness, early warning, and dispatch control of the system. Existing methods mostly use classical mathematics and traditional machine learning to analyze PMU data, which makes it difficult to identify the pattern changes of the data under multiple operating conditions in power systems. In this paper, we propose a hybrid model named IRFLMDNN, which consists of an improved CART random forest model and a dynamic neural network optimized by the Levenberg Marquardt algorithm for PMU data anomaly detection and adaptive data re-filling, respectively. Experimental results based on the IEEE 39-node 10-machine New England Power System show that the proposed method has accurate and robust anomaly detection and data refilling performance. [ABSTRACT FROM AUTHOR]

Parkinson's disease (PD) is a common neurodegenerative disorder that causes degeneration of dopaminergic neurons in the Nigrostriatal pathway and the discharge of Dopamine in the striatum. Machine learning algorithms have been used as a tool to predict and diagnose diseases. Some of these algorithms got the popularity due to their high recognition performance. In the recognition of PD, studies demonstrated various recognition performances and this systematic review study the performance of machine learning algorithms. This systematic review is based on the cochrane's proposed seven phases of review. After identifying the question of research and inclusion/exclusion criteria, we searched different related databases (SID, MagIran, PubMed, ProQuest, ScienceDirect, WoS, Scopus, and Google Scholar) with the help of combination of keywords. After selection of the studies we extract information and summarize the results. From 10,980 found-studies, and removing them based on inclusion/exclustion criteria, we selected 82 studies. To diagnose PD, 59 studies used clinical indicators, 2 studies used genetic characteristics, 12 used MRI, two used PET, 5 used SPECT and 2 used Laboratory markers. In most of these studies RF, SVM, LR have performed the best. The accuracies of RF, SVM, and LR are reported between 58.9%-99.42%, 65.2%-99.99%, and 43.9%-96%. The results show that the performance of RF, SVM, and LR are high for PD diagnosis. Therefore, they can be used for PD diagnosis as a help for doctors and specialist. [ABSTRACT FROM AUTHOR]

Deception is a well-known term that involves acting in a way that it leads another person to believe something that is not true. Deception is a very common occurrence in daily life, and many times it becomes a big problem for national security. To cope with this problem, deception detection has gained a lot of attention recently. In this paper, we have tried to come up with a deceit identification system where we have used electroencephalograph (EEG) data collected by performing a concealed information test. To improve the performance of the system, first we have tried to select the optimal subset of the EEG channels using binary particle swarm optimization and secondly performed support vector machine hyper-parameter optimization using continuous version of PSO. The proposed model is validated using EEG dataset. The performance of the proposed system has resulted in increase of accuracy from 76.98 to 96.45% which is a significant improvement. Also, the proposed approach outperformed in terms of sensitivity, specificity, F1-score and G-measure when compared with state-of-the-art models. [ABSTRACT FROM AUTHOR]

Improving the reliability and performance of electric and rotating machines is crucial to many industrial applications. This will lead to improved robustness, efficiency, and eco-sustainability, as well as mitigate significant health and safety concerns regarding sudden catastrophic failure modes. Bearing degradation is the most significant cause of machine failure and has been reported to cause up to 75% of low-voltage machine failures. This paper introduces a low complexity machine learning (ML) approach to estimate the remaining useful life (RUL) of rolling element bearings using real vibration signals. This work explores different ML recipes using novel feature engineering coupled with various k-Nearest Neighbour (k-NN), and Support Vector Machines (SVM) kernel and weighting functions in order to optimise this RUL approach. Original non-linear wear state models and feature sets are investigated, the latter are derived from Short-time Fourier Transform (STFT) and Hilbert Marginal Spectrum (HMS). These feature sets incorporate one-third octave band filtering for low complexity multivariate feature subspace compression. Our proposed ML algorithm stage has employed two robust supervised ML approaches: weighted k-NN and SVM. Real vibration data were drawn from the Pronostia platform to test and validate this prognostic monitoring approach. The results clearly demonstrate the effectiveness of this approach, with classification accuracy results of up to 82.8% achieved. This work contributes to the field by introducing a robust and computationally inexpensive method for accurate monitoring of machine health using low-cost vibration-based sensing. [ABSTRACT FROM AUTHOR]

Our aim is to investigate the added value of automated machine learning (AutoML) for potential future applications in cancer diagnostics. Using two important diagnostic questions, the non-invasive determination of IDH mutation status and ATRX status, we analyze whether it is possible to use AutoML to develop models that are comparable in performance to conventional machine learning models (ML) developed by experts. For this purpose, we develop AutoML models using different feature preselection methods and compare the results with previously developed conventional ML models. The cohort used for our study comprises T2-weighted MRI images of 124 patients with histologically confirmed gliomas. Using AutoML, we were able to develop sophisticated models in a very short time with only a few lines of computer code. In predicting IDH mutation status, we obtained a mean AUC of 0.7400 and a mean AUPRC of 0.8582. ATRX mutation status was predicted with very similar discriminatory power, with a mean AUC of 0.7810 and a mean AUPRC of 0.8511. In both cases, AutoML was even able to achieve a discriminatory power slightly above that of the respective conventionally developed models in a very short computing time, thus making such methods accessible to non-experts in the near future. [ABSTRACT FROM AUTHOR]

In traditional multi-axis machining, the linear tool path generated by computer-aided manufacturing systems is still in a broad tool path format, resulting in discontinuities at the corners, which slow down tool motion, and also trigger tool vibration, reducing machining accuracy and surface quality. In order to ensure the C3 continuity of the tool during the multi-axis machining process, this paper proposes an analytical B-spline local smoothing method with curvature control. The tool position trajectory and tool orientation are both smoothed by a 5th spline curve with 9 control points. Optimal control points are determined for position and orientation splines to achieve C3 continuity at the junctions between the spline and linear segment while preserving the predefined geometric tolerance limits and minimizing the maximum curvature of the inserted B-spline curve. The smoothed tool path is expected to have less curvature fluctuation, which can significantly improve both the kinematic performance of the machine tool and the machining efficiency. Both computer simulations and physical printing experiments are carried out to demonstrate the improvements of our proposed method in motion smoothness and tracking accuracy in multi-axis machining tasks. [ABSTRACT FROM AUTHOR]

This paper proposes a path-tracking method for agricultural vehicles based on the 4WIS-4WID structure and fuzzy control to improve the operation performance of agricultural machinery in the greenhouse. The influential model of two critical parameters, α and R, on the position correction is obtained based on the relationship analysis between the traditional pure pursuit model and the vehicle structure. Based on aiming pursuit, the relationship equation between the position deviation, lateral deviation d, and heading deviation θ is established. A two-input and two-output fuzzy controller is designed, and the lateral deviation d and heading deviation θ are the input variables. Values of α and R are obtained after fuzzification, fuzzy inference, and defuzzification, and a validated MATLAB model is used to simulate different scenarios. Results of the tests show that the steady-state error of path tracking based on fuzzy control pursuit is between 35 and 51 mm, and the stability distance is between 1661 and 3052 mm for straight path tracking in four initial states. The rectangular corners have the highest inaccuracy. The results of fuzzy control pursuit show a significant improvement in path-tracking performance that can influence vehicle navigation capability in the greenhouse. [ABSTRACT FROM AUTHOR]

Featured Application: The results have been applied to the rice combine harvester produced by China Jiangxi Liangtian Agricultural Machinery Co., Ltd., in China, Thailand, Singapore, and other places to promote the application. The threshing device is the core component of the grain combine harvester, and the straw guide board plays an important role in the threshing device. In the past, the guiding structure of the threshing device could not optimize the working performance of the machine by adjusting the spiral angle. In this study, an adjustable straw guide board was designed, and the movement model of the straw on the straw guide board was analyzed. The response surface method was used to perform field experiments, and the experimental data were analyzed using quadratic polynomial regression. The results show that the drum rotation speed, operating speed, and spiral angle of the straw guide board have significant effects on the percentage of loss rate (PLR), percentage of impurities rate (PIR), and percentage of broken rate (PBR). Further optimization analysis showed that the predicted values of the PLR, PIR, and PBR were 1.18%, 0.72%, and 0.54%, respectively, whereas the experimental verification values were 1.26%, 0.73%, and 0.61%, respectively. The absolute errors between the experimental and predicted values were very small; however, the optimized field test verified values decreased by 8.31%, 50.04%, and 60.30%, respectively, which indicates that the optimized harvester had better operation quality. [ABSTRACT FROM AUTHOR]

The city of Jeddah experienced a severe flood in 2020, resulting in loss of life and damage to property. In such scenarios, a flood forecasting model can play a crucial role in predicting flood events and minimizing their impact on communities. The proposed study aims to evaluate the performance of machine learning algorithms in predicting floods and non-flood regions, including Gradient Boosting, Extreme Gradient Boosting, AdaBoosting Gradient, Random Forest, and the Light Gradient Boosting Machine, using the dataset from Jeddah City, Saudi Arabia. This study identified fourteen continuous parameters and various classification variables to assess the correlation between these variables and flooding incidents in the analyzed region. The performance of the proposed algorithms was measured using classification matrices and regression matrices. The highest accuracy (86%) was achieved by the Random Forest classifier, and the lowest error rate (0.06) was found with the Gradient Boosting regressor machine. The performance of other algorithms was also exceptional compared to existing literature. The results of the study suggest that the application of these machine learning algorithms can significantly enhance flood prediction accuracy, enabling various industries and sectors to make more informed decisions. [ABSTRACT FROM AUTHOR]

To evaluate whether a machine learning classifier can evaluate image quality of maximum intensity projection (MIP) images from F18-FDG-PET scans. A total of 400 MIP images from F18-FDG-PET with simulated decreasing acquisition time (120 s, 90 s, 60 s, 30 s and 15 s per bed-position) using block sequential regularized expectation maximization (BSREM) with a beta-value of 450 and 600 were created. A machine learning classifier was fed with 283 images rated "sufficient image quality" and 117 images rated "insufficient image quality". The classification performance of the machine learning classifier was assessed by calculating sensitivity, specificity, and area under the receiver operating characteristics curve (AUC) using reader-based classification as the target. Classification performance of the machine learning classifier was AUC 0.978 for BSREM beta 450 and 0.967 for BSREM beta 600. The algorithm showed a sensitivity of 89% and 94% and a specificity of 94% and 94% for the reconstruction BSREM 450 and 600, respectively. Automated assessment of image quality from F18-FDG-PET images using a machine learning classifier provides equivalent performance to manual assessment by experienced radiologists. [ABSTRACT FROM AUTHOR]

The stock forecast is one of the most challenging tasks that have attracted numerous economists and scientists worldwide. Stock prices can be affected by many reasons, such as physiological, rational, and irrational behavior. Such factors can combine to make the prices volatile and very challenge to predict with great accuracy for a long time in numerous cases. In this study, we have deployed a long short‐term memory architecture with various time‐steps and classic machine learning methods such as random forests, support vector machines, and autoregression on the Taiwanese stock market collected in 14–15 years. As shown from the visual results, the predicted values have followed the patterns as the actual prices with low error rates in various metrics, including root mean square error and mean absolute error. This work is expected to provide a valuable tool for investigating stock price patterns of stock markets in the future. [ABSTRACT FROM AUTHOR]

This paper proposes a deep learning model that is robust and capable of handling highly uncertain inputs. The model is divided into three phases: creating a dataset, creating a neural network based on the dataset, and retraining the neural network to handle unpredictable inputs. The model utilizes entropy values and a non-dominant sorting algorithm to identify the candidate with the highest entropy value from the dataset. This is followed by merging the training set with adversarial samples, where a mini-batch of the merged dataset is used to update the dense network parameters. This method can improve the performance of machine learning models, categorization of radiographic images, risk of misdiagnosis in medical imaging, and accuracy of medical diagnoses. To evaluate the efficacy of the proposed model, two datasets, MNIST and COVID, were used with pixel values and without transfer learning. The results showed an increase of accuracy from 0.85 to 0.88 for MNIST and from 0.83 to 0.85 for COVID, which suggests that the model successfully classified images from both datasets without using transfer learning techniques. [ABSTRACT FROM AUTHOR]

Hyperparameter optimization plays a significant role in the overall performance of machine learning algorithms. However, the computational cost of algorithm evaluation can be extremely high for complex algorithm or large dataset. In this paper, we propose a model-based reinforcement learning with experience variable and meta-learning optimization method to speed up the training process of hyperparameter optimization. Specifically, an RL agent is employed to select hyperparameters and treat the k-fold cross-validation result as a reward signal to update the agent. To guide the agent's policy update, we design an embedding representation called "experience variable" and dynamically update it during the training process. Besides, we employ a predictive model to predict the performance of machine learning algorithm with the selected hyperparameters and limit the model rollout in short horizon to reduce the impact of the inaccuracy of the model. Finally, we use the meta-learning technique to pre-train the model for fast adapting to a new task. To prove the advantages of our method, we conduct experiments on 25 real HPO tasks and the experimental results show that with the limited computational resources, the proposed method outperforms the state-of-the-art Bayesian methods and evolution method. [ABSTRACT FROM AUTHOR]

Digital twins can improve the level of control over physical entities and help manage complex systems by integrating a range of technologies. The autonomous agricultural machine has shown revolutionary effects on labor reduction and utilization rate in field works. Autonomous vehicles in precision agriculture have the potential to improve competitiveness compared to current crop production methods and have become a research hotspot. However, the development time and resources required in experiments have limited the research in this area. Simulation tools in unmanned farming that are required to enable more efficient, reliable, and safe autonomy are increasingly demanding. Inspired by the recent development of an open-source virtual simulation platform, this study proposed an autoware-based simulator to evaluate the performance of agricultural machine guidance based on digital twins. Oblique photogrammetry using drones is used to construct threedimensional maps of fields at the same scale as reality. A communication format suitable for agricultural machines was developed for data input and output, along with an inter-node communication methodology. The conversion, publishing, and maintenance of multiple coordinate systems were completed based on ROS (Robot Operating System). Coverage path planning was performed using hybrid curves based on Bézier curves, and it was tested in both a simulation environment and actual fields with the aid of Pure Pursuit algorithms and PID controllers. [ABSTRACT FROM AUTHOR]

Accurate prediction of machine RUL plays a crucial role in reducing human casualties and economic losses, which is of significance. The ability to handle spatiotemporal information contributes to improving the prediction performance of machine RUL. However, most existing models for spatiotemporal information processing are not only complex in structure but also lack adaptive feature extraction capabilities. Therefore, a lightweight operator with adaptive spatiotemporal information extraction ability named Involution GRU (Inv-GRU) is proposed for aero-engine RUL prediction. Involution, the adaptive feature extraction operator, is replaced by the information connection in the gated recurrent unit to achieve adaptively spatiotemporal information extraction and reduce the parameters. Thus, Inv-GRU can well extract the degradation information of the aero-engine. Then, for the RUL prediction task, the Inv-GRU-based deep learning (DL) framework is firstly constructed, where features extracted by Inv-GRU and several human-made features are separately processed to generate health indicators (HIs) from multi-raw data of aero-engines. Finally, fully connected layers are adopted to reduce the dimension and regress RUL based on the generated HIs. By applying the Inv-GRU-based DL framework to the Commercial Modular Aero Propulsion System Simulation (C-MAPSS) datasets, successful predictions of aero-engines RUL have been achieved. Quantitative comparative experiments have demonstrated the advantage of the proposed method over other approaches in terms of both RUL prediction accuracy and computational burden. [ABSTRACT FROM AUTHOR]

Hyperspectral imaging (HSI) has the potential to enhance surgical tissue detection and diagnostics. Definite utilization of intraoperative HSI guidance demands validated machine learning and public datasets that currently do not exist. Moreover, current imaging conventions are dispersed, and evidence-based paradigms for neurosurgical HSI have not been declared. We presented the rationale and a detailed clinical paradigm for establishing microneurosurgical HSI guidance. In addition, a systematic literature review was conducted to summarize the current indications and performance of neurosurgical HSI systems, with an emphasis on machine learning-based methods. The published data comprised a few case series or case reports aiming to classify tissues during glioma operations. For a multitissue classification problem, the highest overall accuracy of 80% was obtained using deep learning. Our HSI system was capable of intraoperative data acquisition and visualization with minimal disturbance to glioma surgery. In a limited number of publications, neurosurgical HSI has demonstrated unique capabilities in contrast to the established imaging techniques. Multidisciplinary work is required to establish communicable HSI standards and clinical impact. Our HSI paradigm endorses systematic intraoperative HSI data collection, which aims to facilitate the related standards, medical device regulations, and value-based medical imaging systems. [ABSTRACT FROM AUTHOR]

Medical data processing and analysis using machine learning algorithms are a prominent research topic these days. To obtain high performances, most state-of-the-art models must be trained on a large number of labeled datasets. However, manually collecting a large-scale real-world medical data is expensive due to the issues like privacy, security, and reliability. Moreover, the collected samples may be incomplete, i.e., some important data items are missing, which has a significant impact on the performance of the machine learning algorithms, especially deep learning models. In this paper, we present a strategy that uses the idea of adversarial learning to augment the real-world medical samples with the incomplete issue to obtain better performance in predicting patient states. Rather than supplying the incomplete data sample with extra instance-level information as in existing data-complementation methods, our method aims to achieve the complementary effect in the feature level without consuming human effort. The method receives both high-quality data and low-quality data (with serious incomplete issue) to learn a comprehensive feature space where the incomplete samples can be complemented with the knowledge transferred from the high-quality samples by the adversarial learning scheme. From the perspective of feature representations, our method can alleviate the incomplete issue of the low-quality data, which enhances the model performance in the end task. Experiments show that our method works well on a real-world medical dataset, which collected for assessment of pelvic organ prolapse. When compared to machine learning approaches frequently employed in medical data analysis, our approach shows a significant improvement of up to 10%. [ABSTRACT FROM AUTHOR]