Search

The deterioration of fiber properties during recycling processes, especially the loss of tensile strength, raises concerns that paper products made from recycled fibers might not satisfy quality requirements. The purpose of this paper is to estimate the deterioration of tensile strength and the damage in paper sheets made of recycled fibers using the theory of damage mechanics and machine learning methods. Experiments were carried out to recycle wood fibers and nonwood fibers four times, and the physicochemical properties of the handsheets made from these fibers were measured after each recycling. Water retention value and relative bonded area were selected as the features to estimate and predict tensile strength during recycling because they had strong correlations with tensile strength. This paper proposed a damage index to quantitatively express the severity of the damage in paper sheets based on the experimental investigation and the theory of damage mechanics. Thus, the deterioration of tensile strength could be estimated and predicted. To determine the damage index, a curve fitting model based on the hyperbolic theory of pulp properties was developed. The proposed quantitative expression of the damage index is: D = (Ds - h)²/a2 - (Dh - k)²/b2, where the coefficients were determined through the curve fitting model. This paper also developed a long short-term memory recurrent neural network model to determine the damage index according to the sequence of recycling. Both models were trained with the experimental data of water retention value and relative bonded area. The estimation and prediction by the curve fitting model were more accurate than those of the neural network model. The root mean square errors by the curve fitting model were 0.0278 for estimation, 0.1667 for prediction; and by the neural network model were 0.2445 for estimation, 0.2206 for prediction, respectively. After the damage index was determined, the deterioration of tensile strength then could be calculated as T = T0 (1-D). [ABSTRACT FROM AUTHOR]

With the increasing integration of AI tools like Yewno Discover, Scholarcy, and Grammarly in academic libraries, undergraduate research has witnessed transformative changes. These tools, while elevating the research process, also bring forth challenges rooted in ethics and application. This paper explores the synergy between modern technology and academic exploration, highlighting the benefits and potential pitfalls of using AI in the research workflow. It emphasizes that while Yewno Discover and similar tools offer streamlined navigation of vast information databases, it is imperative for undergraduates to remain cognizant of potential biases and other ethical considerations. This paper underscores the need for proactive measures in academic settings, including specialized training and policy development, to ensure that undergraduate researchers harness the power of AI responsibly and efficiently. [ABSTRACT FROM AUTHOR]

This Special Issue presents 7 papers selected from the International Conference on Futuristic Advancements in Materials, Manufacturing and Thermal Sciences (ICFAMMT 2022). These papers were presented during the International Conference on Futuristic Advancements in Materials, Manufacturing and Thermal Sciences (ICFAMMT 2022) and selected by Session Chairs and Scientific Committee members for the possible publication in the Special Issue of Heat Transfer Engineering. In this special issue, the first paper by Kumar et al. presents heat transfer characteristics of plate heat exchanger with bubble fin using Al SB 2 sb O SB 3 sb /water nanofluid. [Extracted from the article]

This paper focuses on the opportunities and challenges associated with the use of artificial intelligence (AI) in academic library operations. In the quest to render fast, effective and efficient services, academic libraries have adopted different technologies in the past. Artificial intelligence technologies is the latest among the technologies currently being introduced in libraries. The technology which is considered an intelligent system, come in the form of robots and expert systems which have natural language processing, machine learning and pattern recognition capabilities. This paper examined the features of AI, the application of AI to library operations, examples of academic libraries with AI technologies in Sub-Saharan Africa, the need for AI in libraries and the challenges associated with the adoption of AI in libraries. The study concluded that AI holds a lot of prospects for the improvement of information services delivery in African academic libraries. Consequently, its adoption is a sinequanon to delivering robust library services in the Fourth Industrial Revolution (4IR). [ABSTRACT FROM AUTHOR]

This paper highlights the importance of involving academic librarians in the development of ethical AI policies. The Academic Librarian Framework for Ethical AI Policy Development (ALF Framework) is introduced, recognizing librarians' unique skills and expertise. The paper discusses the benefits of their involvement, including expertise in information ethics and privacy, practical experience with AI tools, and collaborations. It also addresses challenges, such as limited awareness, institutional resistance, resource constraints, interdisciplinary collaboration, and evolving AI technologies, offering practical solutions. By actively involving librarians, institutions can develop comprehensive and ethical AI policies that prioritize social responsibility and respect for human rights. [ABSTRACT FROM AUTHOR]

Recognition of spatio-temporal traffic patterns at the network-wide level plays an important role in data-driven intelligent transport systems (ITS) and is a basis for applications such as short-term prediction and scenario-based traffic management. Common practice in the transport literature is to rely on well-known general unsupervised machine-learning methods (e.g., k-means, hierarchical, spectral, DBSCAN) to select the most representative structure and number of day-types based solely on internal evaluation indices. These are easy to calculate but are limited since they only use information in the clustered dataset itself. In addition, the quality of clustering should ideally be demonstrated by external validation criteria, by expert assessment or the performance in its intended application. The main contribution of this paper is to test and compare the common practice of internal validation with external validation criteria represented by the application to short-term prediction, which also serves as a proxy for more general traffic management applications. When compared to external evaluation using short-term prediction, internal evaluation methods have a tendency to underestimate the number of representative day-types needed for the application. Additionally, the paper investigates the impact of using dimensionality reduction. By using just 0.1% of the original dataset dimensions, very similar clustering and prediction performance can be achieved, with up to 20 times lower computational costs, depending on the clustering method. K-means and agglomerative clustering may be the most scalable methods, using up to 60 times fewer computational resources for very similar prediction performance to the p-median clustering. [ABSTRACT FROM AUTHOR]

Prediction of heart diseases on time is significant in order to preserve life. Many conventional methods have taken efforts on earlier prediction but faced with challenges of higher prediction cost, extended time for computation and complexities with larger volume of data which reduced prediction accuracy. In order to overcome such pitfalls, AI (Artificial Intelligence) technology has been evolved in diagnosing heart diseases through deployment of several ML (Machine Learning) and DL (Deep Learning) algorithms. It improves detection by influencing with its capacity of learning from the massive data containing age, obesity, hypertension and other risk factors of patients and extract it accordingly to differentiate on the circumstances. Moreover, storage of larger data with AI greatly assists in analysing the occurrence of the disease from past historical data. Hence, this paper intends to provide a review on different AI based algorithms used in the heart disease prognostication and delivers its benefits through researching on various existing works. It performs comparative analysis and critical assessment as encompassing accuracies and maximum utilization of algorithms focussed by traditional studies in this area. The major findings of the paper emphasized on the evolution and continuous explorations of AI techniques for heart disease prediction and the future researchers aims in determining the dimensions that have attained high and low prediction accuracies on which appropriate research works can be performed. Finally, future research is included to offer new stimulus for further investigation of AI in cardiac disease diagnosis. [ABSTRACT FROM AUTHOR]

As a vital component of mining hoisting equipment, mining wire rope (MWR) is critical in the operation of the mine. Once damaged, it is very easy to cause the loss of life and property. Therefore, it is meaningful to identify the damage to MWR, especially the weak damage identification. This paper proposes a method based on continuous wavelet transform (CWT) and convolutional neural network (CNN) model for MWR damage degree identification. Firstly, the MWR signals with different damage degrees were acquired. Secondly, CWT and data augmentation were performed on the original signals to obtain a time-frequency image dataset of MWR damage degree. Then, a deep learning (DL) model is built for identification and compared with traditional machine learning and some classical CNN models. The results show that the model proposed in this paper is better than the traditional machine learning models. It has the highest accuracy of 91.58% in MWR damage degree identification with classical CNN models. The method focuses on the weak damage of MWR and combines CWT with CNN model to meet the requirements of accuracy and efficiency, detecting the early damage as soon as possible and ensuring intelligent, safe and stable operation in the mining industry. [ABSTRACT FROM AUTHOR]

AbstractWidespread innovation from artificial intelligence and machine learning (AI/ML) tools presents a lucrative opportunity for the nuclear industry to improve state-of-the-art analyses (e.g. condition monitoring, remote operation, etc.) due to increased data visibility. In recent years, the risk posed by collaborative data exchange has received increased attention due, in part, to a potential adversary’s ability to reverse-engineer intercepted data using domain knowledge and AI/ML tools. While the efficacy of typical encryption has been proven during passive communication and data storage, collaborative exchange typically requires decryption for extended analyses by a third-party, which poses an intrinsic risk due to these trustworthiness concerns. The directed infusion of data (DIOD)11 A.Al Rashdan and H.Abdel-Khalik, Deceptive Infusion of Data, Non-Provisional Patent, Application No. 63/227,389, September 2022. paradigm presented in this paper discusses a novel data masking technique that relies on preserving the usable information of proprietary data while concealing its identity via reduced-order modeling. In contrast to existing state-of-the-art data masking methods, DIOD does not impose limiting assumptions, computational overheads, or induced uncertainties, thereby allowing for secure and flexible data-level security that does not alter the inferential content of the data. This paper focuses on the application of DIOD to a process-based simulation wherein a leaking reservoir with a controlled inlet pump is simulated under various experimental conditions with the goal of producing masked data that preserve the information given by anomalies. These experiments included the injection of statistically significant anomalies, subtle anomalies that occurred over an extended period, and the addition of several independent anomalous states. Each experiment showed that a classifier will identify the same anomalies whether it analyzes the original or masked data. An additional experiment also tested the case of corrupted labeling information wherein labels were arbitrarily randomized, and the loss in labeling accuracy was about the same for both datasets. Each of these experiments show that data obfuscated by DIOD may be utilized in the place of real data for a variety of condition monitoring scenarios with no loss in performance.A.Al Rashdan and H.Abdel-Khalik, Deceptive Infusion of Data, Non-Provisional Patent, Application No. 63/227,389, September 2022. [ABSTRACT FROM AUTHOR]

With the new revolution in data technology, many types of streaming data are automatically generated in our living environment. The vast amount of information carried by this streaming data demands decision support from new online learning algorithms. In this paper, we propose a new online learning algorithm to monitor changes in stream data/system status and provide automatic decision support using streaming data. The new online learning algorithm we propose consists of two components. Firstly, we adapt an existing Bayesian approach into an online learning algorithm for change detection, which serves as Component A. Then, we integrate this algorithm with a modified online learning method from repeated games (Play against Random Past), which forms Component B. Theoretical support for Component B is provided with a mathematical proof at the end. We demonstrate the algorithm's performance in this paper through simulations using both artificial data from a random process and data from the 2009 H1N1 pandemic. [ABSTRACT FROM AUTHOR]

Transfer learning is a method to improve the estimation and prediction accuracy of the target model by transferring the source data when the available data of the target data is relatively few. However, existing transfer learning methods tend to ignore the heterogeneity and heavy-tailedness of high-dimensional data. So we consider the Huber regression, which is robust to thick-tailed distributed data and outliers. Based on the high-dimensional model, this paper adds the constraint conditions of prior information, carries out the research of robust transfer learning and uses Huber regression to estimate the coefficients of the target model. In this paper, we design corresponding transfer learning algorithms for known and unknown transferable sources, and prove the effectiveness of this method through simulation experiments and actual cases. [ABSTRACT FROM AUTHOR]

This paper interrogates the informational practices shared between human and computer machine learners as they train to sense the world through lines of order, or vectors. The paper does this by exploring the affective conditions through which vectors draw relations of data over a persistent, colonial image of race. Through analysis of pedagogical practices at the Summer Institute for Computational Social Science in Chicago, and a corresponding year-long machine learning design group, this paper examines how contemporary machine learning practitioners train themselves to sense calculative relationality on the basis of racialized difference. The paper compares this vectorized sensibility with 20th century enumerative practices in the United States by analyzing the racial statistics of W.E.B. Du Bois, Kelly Miller, and Frances Kellor to trace out affective histories of the vector. Ultimately, this paper asks how machine learners – whether algorithms or their human users – often project lines of colonial order upon other forms of life, and how, by questioning the claim of vector relations and their informational objects, we can confront this sense-training and reimagine ourselves. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) and its data-driven counterpart, machine learning (ML), are rapidly evolving disciplines with increasing applications in modeling, simulation, control, and optimization within the drying industry. This paper presents a comprehensive overview of progress made in ML from shallow to deep learning and its implications for food drying. Theoretical foundations, advantages, and limitations of various ML approaches employed in this domain are explored. Additionally, advancements in ML models, particularly those enhanced by optimization algorithms, are reviewed. The review underscores the role of intelligent configuration of ML models, which affects their accuracy and ability to solve problems of high energy consumption, nutrient degradation, and uneven drying. Drawing upon research achievements, integrating of AI models with real-time measuring methods is discussed, enabling dynamic determination of optimal drying conditions and parameter adjustments. This integration facilitates automated decision-making, reducing human errors and enhancing operational efficiency in food drying. Moreover, AI models demonstrate proficiency in predicting drying times and analyzing energy usage patterns, thereby enabling optimization to minimize resource consumption while preserving product quality. Finally, this paper identifies current obstacles in technology development and proposes novel research avenues for sustainable drying technologies. The strengths and weaknesses of various AI methodologies are examined Artificial neural networks are extensively used for modeling drying phenomena Machine learning models can simulate complex processes of food drying Deep learning has significant potential for real-time monitoring of drying Intelligent control systems can optimize food drying [ABSTRACT FROM AUTHOR]

This paper addresses a notable gap in the field of photovoltaic system forecasting by introducing the Machine Learning-based PV Prediction and Fault Analysis System (ML-PVPFAS). This framework is designed to optimize the decomposition of variational systems automatically, using a multi-objective intelligent optimization method to establish its weight. The paper evaluates a range of Machine Learning (ML) and Ensemble Learning (EL) techniques for diagnosing faults in PV arrays, with a primary focus on identifying and categorizing complex faults that could affect these arrays. These faults encompass multiple anomalies and defects with similar current-voltage curves, which have not been previously examined. The analysis of prediction scores reveals that the ML-PVPFAS approach outperforms other methods, with the lowest Mean Absolute Percentage Error (MAPE) at 4.93, Root Mean Squared Error (RMSE) at 4.64, high Accuracy at 90.31%, Precision at 93.60%, a strong Pearson Correlation Coefficient of 0.90, and a fast Computation Time of 77.64 milliseconds. The results suggest that ML-PVPFAS is a dependable and practical algorithm for predicting the power output of PV solar systems, making it a valuable contribution to the field of predictive modeling. [ABSTRACT FROM AUTHOR]

AbstractWith the development of intelligent transportation, it has become mainstream for drivers and automated systems to cooperate to complete train driving tasks. Human-machine trust has become one of the biggest challenges in achieving safe and effective human-machine cooperative driving. Accurate evaluation of human-machine trust is of great significance to calibrate human-machine trust, realize trust management, reduce safety accidents caused by trust bias, and achieve performance and safety goals. Based on typical driving scenarios of high-speed trains, this paper designs a train fault judgment experiment. By adjusting the machine’s reliability, the driver’s trust is cultivated to form their cognition of the machine. When the driver’s cognition is stable, data from the Trust in Automated (TIA) scale and modes of physiological information, including electrodermal activity (EDA), electrocardiograms (ECG), respiration (RSP), and functional near-infrared spectroscopy (fNIRS), are collected during the fault judgment experiment. Based on analysis of this multi-modal physiological information, a human-machine trust classification model for high-speed train drivers is proposed. The results show that when all four modes of physiological information are used as input, the random forest classification model is most accurate, reaching 93.14%. This indicates that the human-machine trust level of the driver can be accurately represented by physiological information, thus inputting the driver’s physiological information into the classification model outputs their level of human-machine trust. The human-machine trust classification model of high-speed train drivers built in this paper based on multi-modal physiological information establishes the corresponding relationship between physiological trust and human-machine trust level. Human-machine trust level is characterized by physiological trust monitoring, which provides support for the dynamic management of trust. [ABSTRACT FROM AUTHOR]

Despite the complexity of thermodynamic models in predicting wax precipitation, they have not shown satisfactory results in the literature. This paper proposes the least-squares support vector machine optimized by the genetic algorithm to forecast wax precipitation based on the mole fractions, specific gravity, and temperature of crude oil samples. The genetic algorithm was used to prevent sticking in local optima in optimizing the parameters of the predictive model. A dataset of 88 experimental wax precipitation samples was used to evaluate the performance of the proposed method. The results revealed that the suggested algorithm could predict wax precipitation with an accuracy of 0.939 and 0.997 before and after sensitivity analysis, respectively. Furthermore, sensitivity analysis improved the Coefficient of Determination, Mean Squared Error, and Root Mean Squared Error from 0.939, 0.768, and 0.877 to 0.997, 0.041, and 0.202, respectively. The results were compared with the results of previous data-driven and mathematical models that showed the outperformance of the proposed method. The contribution of this paper can be summarized as demonstrating the ability of intelligent models to predict wax precipitation without a need for complex thermodynamic models. Also, it was shown that a proper feature selection could improve the model's performance. [ABSTRACT FROM AUTHOR]

Quantum Machine Learning is an interdisciplinary field that combines the principles of quantum physics, quantum computers, and machine learning to enhance computational performance. In recent years, there have been significant advancements in the field of quantum machine learning. Researchers have developed quantum algorithms for various tasks. Furthermore, quantum machine learning has been applied to various domains, such as chemistry, cryptography, and finance. Previous bibliometric surveys in this area covered the period from 2014 to 2020. However, as demonstrated in this paper, the pace of publication has increased significantly from 2018 to the present. Therefore, it has become essential to provide a new analysis of the current state of research in this field. In this paper, we deployed visualization tools to analyze co-authorships, co-occurrences, and keyword density in this research area. Our study covers and analyzes a total of 918 publications from the Web of Science database and 1171 publications from the Scopus database from 2006 to 2022. Following the analysis, we identify research questions, opportunities, and research gaps in the field of Quantum Machine Learning. [ABSTRACT FROM AUTHOR]

Artificial Intelligence is quickly emerging as a technological solution for the agriculture industry to surmount its classical challenges. Artificial Intelligence is facilitating farmers to refine their products and alleviate unfavourable impacts due to the environment. The central concern of this paper is predictive analytics to develop a machine learning model to identify and predict crop yield based on multiple environmental factors. In this paper, a hybrid learner 'RaNN' is proposed that combines the feature sampling and majority voting technique of Random Forest in-combination with the multilayer Feedforward Neural Network to predict the crop yield. Research has also ascertained the essential features responsible for accurate yield prediction. The proposed model works for rice yield prediction, one of the chief grains of India. The region chosen for the work is Punjab, which is among the largest producer states of India for rice. The dataset consists of 15 attributes comprising the weather and agriculture data collected from the Indian Meteorological Department Pune, and Punjab Environment Information System (ENVIS) Center, Government of India. The study has also made a comparative assessment of 'RaNN' with machine learning methods like Multiple Linear Regression, Random Forest, Decision Tree, Boosting Regression, Support Vector Machine Regression, Ensemble Learner, and Artificial Neural Network. Our model RaNN has listed a better prediction accuracy with minimal error among the other techniques providing a 98% correlation between the actual and the predicted yield. Abbreviations: AI – Artificial Intelligence; ANN – Artificial Neural Network; BR – Boosting Regression; Chem Fert – chemical fertilisers; DT – Decision Tree; EL – Ensemble Learner; ENVIS – Punjab Environment Information System; GBM – Stochastic Gradient Boosting Method; GPS – Global Positioning System; HMAX – highest maximum temperature in degrees C; IMD – Indian Meteorological Department; L1 – Lasso regression; L2 – Ridge regression; LMIN – lowest minimum temperature; ML – Machine Learning; MAE – Mean Absolute Error; MEVP – mean evaporation in mm; MLR – Multiple Linear Regression; MMAX – mean maximum temperature in degrees C; MMIN – mean minimum temperature in degrees C; MSSH – Mean sunshine duration in hours; MWS – mean wind speed in km/h; P1 – number of days with precipitation (0.1–0.2 mm); P2 – number of days with precipitation (greater than or equal to 0.3 mm); RaNN – Hybrid RF-ANN model; RMSE – Root Mean Squared Error; $${R^2}$$ R 2 – Coefficient of determination; RD – number of rainy days; RF – Random Forest; SVM Reg – Support Vector Machine Regression; TMRF – total rainfall per month in mm [ABSTRACT FROM AUTHOR]

PurposeMethodologyFindingsOriginalityResearch implicationsPractical implicationsThe general purpose of shipping market prediction is to help shipping analysts design a strategy to make shipping strategies. Previous studies have pointed out human emotions has an obvious impact on markets, however, few research focus on utilizing human emotions to improve the shipping market forecasting issue. This paper fills the gap in the literature of considering industrial individual emotions to shipping market prediction issueWe proposed a conceptual framework of an emotion-based shipping market prediction (ESMP) system focused on considering the multidimensional emotions of individual investors. To make it precisely, a two-fold Particle Swarm Optimization (TFPSO) algorithm is proposed conduct the shipping forecasting under the framework of ESMP, with advantages of automatically finding the architecture and hyperparameters of the Deep Long Short-Term Memory (DLSTM) simultaneously in the shipping market forecasting the values of the Baltic Dry, Dirty Tanker and Container indices of ocean transportation in the world ocean during the crisis period 2010–2022The prediction accuracy of the ESMP was higher than that of the models using the conventional factors in all learning periods of the study, and it was determined that the results complement the sentiment indicator employed to predict the shipping indices. Additionally, the accuracy of the proposed method is superior to conventional neural network models in all used error metrics. Additional Mann–Whitney U test on MSE difference between TFPSO-DLSTM and compared models demonstrates that the significant advantages boosted by TFPSO-DLSTMShipping sentiment has provided further proof that shipping decisions are significantly driven by emotions. our study shows that individual shipping investors’ sentiment are able to improve shipping rate forecasts significantly, although the magnitudes of the improvements are relatively small from an economic point of viewThis study undertook the initiative to procure BDI, BDTI, BCTI, COVID-19 data and shipping sentiment index data through in-depth interviews to provide first-hand perspectives into whether shipping sentiments impact the shipping market trend. It builds upon existing literature on the present stance of deep learning models, which largely relies non-sentiment factors. The study also extends prior literature on hyperparameter searching methods by highlighting the structure of searching method. Besides this, the contribution of this study also aligns with the prior discussions on evolving methodologies in temporal forecasting researchThis study implies that shipping sentiments enables to make better shipping marketing trends, and foster sustainable growth. Shipping fear index serves for measuring investors’ attitude and mood toward shipping markets in terms of the general and certain sectors or assets, possibly promoting the movement of price and providing long-term investors and active traders with trading or arbitrage opportunities. Due to the complexities and regulations of the shipping market, it is essential to collaborate exclusively with shipping sentiment indices with artificial intelligent model in shipping market forecasting research [ABSTRACT FROM AUTHOR]

AbstractIn this paper, we study the transfer learning problem in functional classification, aiming to improve the classification accuracy of the target data by leveraging information from related source datasets. To facilitate transfer learning, we propose a novel transferability function tailored for classification problems, enabling a more accurate evaluation of the similarity between source and target dataset distributions. Interestingly, we find that a source dataset can offer more substantial benefits under certain conditions than another dataset with an identical distribution to the target dataset. This observation renders the commonly-used debiasing step in the parameter-based transfer learning algorithm unnecessary under some circumstances to the classification problem. In particular, we propose two adaptive transfer learning algorithms based on the functional Distance Weighted Discrimination (DWD) classifier for scenarios with and without prior knowledge regarding informative sources. Furthermore, we establish the upper bound on the excess risk of the proposed classifiers, providing the statistical gain via transfer learning mathematically provable. Simulation studies are conducted to thoroughly examine the finite-sample performance of the proposed algorithms. Finally, we implement the proposed method to Beijing air-quality data, and significantly improve the prediction of the PM 2.5 level of a target station by effectively incorporating information from source datasets. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel semi-analytic physics informed neural network (PINN) method for solving singularly perturbed boundary value problems. The PINN is a scientific machine learning framework that shows great promise for finding approximate solutions to partial differential equations. PINNs have demonstrated impressive performance in solving a variety of differential equations, including time-dependent and multi-dimensional equations involving complex domain geometries. However, when it comes to stiff differential equations, neural networks in general struggle to capture the sharp transition of solutions, due to the spectral bias. To address this limitation, we develop a semi-analytic PINN approach, which is enriched by incorporating the so-called corrector functions obtained from boundary layer analysis. Our enriched PINN approach provides accurate predictions of solutions to singular perturbation problems. Our numerical experiments cover a wide range of singularly perturbed linear and nonlinear differential equations. Overall, our approach shows great potential for solving challenging problems in the field of partial differential equations and machine learning. [ABSTRACT FROM AUTHOR]

Automated employment decision tools use machine learning, artificial intelligence, predictive analytics, and other data-driven approaches to enhance candidate experiences and streamline employment related decision-making, allowing human resources to be concentrated where they are needed most. However, the use of these tools without appropriate safeguards has resulted in a number of high-profile scandals in recent years, particularly in regard to bias. Accordingly, lawmakers have started to propose laws that require bias audits of automated employment decision tools to examine their outputs for subgroup differences. The first of its kind was New York City Local Law 144, but other US states have since followed suit. In this paper, we examine the concerns about the effectiveness of this and other similar laws, including the suitability of metrics, the scope of the law, and low levels of compliance. We conclude that despite the law being a good initial first step towards greater transparency around automated employment decision tools and reducing bias, examining outcomes alone is not sufficient to prevent bias elsewhere in the tool. Moreover, effective bias prevention will require a multidisciplinary approach that combines expertise in IO psychology, law, and computer science to develop appropriate metrics and maximize the enforceability of such laws. [ABSTRACT FROM AUTHOR]

Byzantine Fault Tolerance (BFT) is crucial for ensuring the resilience of Distributed Machine Learning (DML) systems during training under adversarial conditions. Among the rising corpus of research on BFT in DML, there is no comprehensive classification of techniques or broad analysis of different approaches. This paper provides an in-depth survey of recent advancements in BFT for DML, with a focus on first-order optimisation methods, particularly, the popular one Stochastic Gradient Descent (SGD) during the training phase. We offer a novel classification of BFT approaches based on characteristics such as the communication process, optimisation method, and topology setting. This classification aims to enhance the understanding of various BFT methods and guide future research in addressing open challenges in the field. This work provides the foundations for developing robust BFT systems, using a variety of optimisation methods to strengthen resilience. [ABSTRACT FROM AUTHOR]

AbstractBased on the sound speed and physical property measurements of seafloor sediments obtained in the central part of the South Yellow Sea, a multi-parameter prediction model for the sound speed of seafloor sediments based on seven parameters, including density, porosity, water content, liquid limit, plasticity index, compression coefficient, and median grain size, was developed by using the BP-AdaBoost fusion algorithm. The results show that the model, configured with 7 input layer neurons, 10 hidden layer neurons, 1 output node, a learning rate of 0.1, and 300 training iterations, achieved a correlation coefficient of 0.962. The mean absolute error (MAE) of the predicted sound speed was 10.19 m/s, and the mean relative error (MAPE) was 0.66%. These results are better than those of single-parameter and dual-parameter prediction equations and other machine learning models. The BP-AdaBoost model prediction method of sound speed of seafloor sediments established in this paper is better than the single-parameter and dual-parameter prediction equations and other prediction models, and it can provide a new way for the prediction of sound speed of seafloor sediments in the study area. [ABSTRACT FROM AUTHOR]

Flotation is a crucial separation technique in coal beneficiation. However, the inability to quickly and accurately detect ash content during the flotation process hinders the development of flotation automation. This paper introduced a novel hybrid model based on convolutional neural networks, establishing the ash content in tailings. In this method, convolutional neural network is utilized to automatically extract features from tailings images. In the model training and optimization phase after inputting the augmented data, the optimal ResNet18 model is identified by comparing five model optimization methods and replacing the SoftMax classifier with the traditional SVR in ResNet18. Additionally, a PCA-SVR model for predicting tailings ash content is constructed for comparative analysis, employing six different feature combinations to validate the effectiveness of the proposed features, with optimal support vector parameters identified through random search and cross-validation algorithms. The prediction results show that the PCA-SVR model has the highest accuracy when the feature dimension is reduced to 16, and the prediction of ResNet18-SVR is more accurate compared to PCA-SVR. This indicates that the proposed method can predict the ash content more accurately than the traditional feature engineering methods and has a broad application prospect in the field of coal beneficiation. [ABSTRACT FROM AUTHOR]

AbstractObjectivesPatients and methodsResultsConclusionNeuropsychological tests (NPTs) are standard tools for assessing cognitive function. These tools can evaluate the cognitive status of a subject, which can be time-consuming and expensive for interpretation. Therefore, this paper aimed to optimize the systematic NPTs by machine learning and develop new classification models for differentiating healthy controls (HC), mild cognitive impairment, and Alzheimer’s disease dementia (ADD) among groups of subjects.A total dataset of 14,926 subjects was obtained from the formal 46 NPTs based on the Seoul Neuropsychological Screening Battery (SNSB). The statistical values of the dataset included an age of 70.18 ± 7.13 with an education level of 8.18 ± 5.50 and a diagnosis group of three; HC, MCI, and ADD. The dataset was preprocessed and classified in two- and three-way machine-learning classification from scikit-learn (www.scikit-learn.org) to differentiate between HC versus MCI, HC versus ADD, HC versus Cognitive Impairment (CI) (MCI + ADD), and HC versus MCI versus ADD. We compared the performance of seven machine learning algorithms, including Naïve Bayes (NB), random forest (RF), decision tree (DT), k-nearest neighbors (KNN), support vector machine (SVM), AdaBoost, and linear discriminant analysis (LDA). The accuracy, sensitivity, specificity, positive predicted value (PPV), negative predictive value (NPV), area under the curve (AUC), confusion matrixes, and receiver operating characteristic (ROC) were obtained from each model based on the test dataset.The trained models based on 29 best-selected NPT features were evaluated, the model with the RF algorithm yielded the best accuracy, sensitivity, specificity, PPV, NPV, and AUC in all four models: HC versus MCI was 98%, 98%, 97%, 98%, 97%, and 99%; HC versus ADD was 98%, 99%, 96%, 97%, 98%, and 99%; HC versus CI was 97%, 99%, 92%, 97%, 97%, and 99% and HC versus MCI versus ADD was 97%, 96%, 98%, 97%, 98%, and 99%, respectively, in predicting of cognitive impairment among subjects.According to the results, the RF algorithm was the best classification model for both two- and three-way classification among the seven algorithms trained on an imbalanced NPTs SNSB dataset. The trained models proved useful for diagnosing MCI and ADD in patients with normal NPTs. These models can optimize cognitive evaluation, enhance diagnostic accuracy, and reduce missed diagnoses. [ABSTRACT FROM AUTHOR]

Rayleigh waves are highly sensitive to surface and subsurface defects. However, the varying surface profiles of complex structures induce constant changes in the Rayleigh waveform. Furthermore, Rayleigh waves encompass multi-dimensional characteristics of subsurface defects, significantly increasing the difficulty of quantitatively detecting the sizes of subsurface defects in different dimensions. This paper presents a machine learning (ML)-based method for quantitatively detecting subsurface defect depth and width in curved structures using laser-generated Rayleigh waves. An experimentally validated finite element model is first established to construct a dataset of Rayleigh wave signals labeled with multidimensional subsurface defect sizes. The dataset is augmented from 188 to 368 samples using the Mix-up algorithm, with additional experimental signals incorporated. Feature parameters of Rayleigh waves are extracted in both time and frequency domains through waveform analysis and sliding window wavelet transform (SWT). These features are used to train ML models: Gaussian Process (GP), Extreme Gradient Boosting (XGBoost), and Random Forest (RF). Comparative results demonstrate GP achieves the highest prediction accuracy for subsurface defect depth and width on the test set, reaching 98.64% and 97.73%, respectively. The study highlights the integration of experimental and simulated data in training, which reduces costs while enhancing model robustness in practical applications. [ABSTRACT FROM AUTHOR]

Buttenfield (1988) pioneered research on multiple representations in the dawn of GIScience. Her efforts evoked inquiries into fundamental issues arising from the selective abstractions of infinite geographic complexity in spatial databases, cartography, and application needs for varied geographic details. These fundamental issues posed ontological challenges (e.g. entity classification) and implementational complications (e.g. duplication and inconsistency) in geographic information systems (GIS). Expanding upon Buttenfield's line of research over the last three decades, this study reviewed multiple representations in spatial databases, spatial cognition, and deep learning. Initially perceived as a hindrance to GIS, multiple representations were found to offer new perspectives to encode and decipher geographic complexity. This paper commenced by acknowledging Buttenfield's pivotal contributions to multiple representations in GIScience. Subsequent discussions synthesized the literature to outline cognitive representations of space in the brain's hippocampal formation and feature representations in deep learning. By cross-referencing related concepts of multiple representations in GIScience, the brain's spatial cells, and machine learning algorithms, this review concluded that multiple representations facilitate learning geography for both humans and machines. [ABSTRACT FROM AUTHOR]

Air pollution affects not only the air in cities but also extends to all indoor environments (homes, offices, schools, public places, transportation, etc.), where we spend between 80% and 90% of our time. Both indoor and outdoor air quality have emerged as significant health concerns and are integral to national strategies implemented by health and environmental institutes in each country. Recently, complaints regarding outdoor air quality have risen in cities, primarily due to automobile traffic and industrial activities in urban areas, and also indoors within homes, offices, and schools. The following paper presents a methodology for the calibration of low-cost monitoring stations based on measurements in a couple of cities in Colombia as part of the development of a project to reduce the environmental awareness gap in urban areas for the estimation of the air quality through low-cost, flexible, modular, and mobile air quality monitoring station design that could be used to assess air pollution in different indoor and outdoor environments. With the implementation of the low-cost stations, we have calibrated and evaluated the performance of the stations using usual linear regression methods, but we have also explored the use of unsupervised estimation with the help of machine learning algorithms, specifically with Random Forest estimators. We have found a significant improvement with using Random Forest for station calibration compared with those found using simple linear regressions for calibration effects. We have found that all the models offer a significant improvement in terms of RMSE. The regression model improves RMSE by up to 70%, while the multiple regression model does so by up to 73%. However, it is the Random Forest that shows the most remarkable improvement, with a reduction in RMSE of up to 86%. [ABSTRACT FROM AUTHOR]

The aim of this paper is to develop a marginal screening method for variable screening in high-dimensional binary classification based on the Wasserstein distance accounting for the distributional difference. Many existing screening methods, such as the two-sample t-test and Kolmogorov test, have been developed under the parametric/nonparametric modeling assumptions to reduce the dimension of the predictors. However, such modeling specifications or nonparametric approaches are associated with the probability measure induced by the predictor in a Euclidean space. While many machine learning methods have successfully found the nonlinear decision boundary in the transformed space, called the reproducing kernel Hilbert space (RKHS), we consider the Wasserstein filter's capacity to detect the distributional difference between two probability measures induced by the nonlinear function of the predictor in the RKHS. Thereby, we can flexibly filter out the non-informative predictors associated with the binary classification, as well as escape the modeling assumptions required in a Euclidean space. We prove that the Wasserstein filter satisfies the sure screening property under some mild conditions. We also demonstrate the advantages of our proposed approach by comparing the finite sample performance of it with those of the existing choices through simulation studies, as well as through application to lung cancer data. [ABSTRACT FROM AUTHOR]

The ability to estimate future velocity on a road network is relevant for applications such as vehicle navigation systems and emergency vehicle dispatching. The existence of traffic congestion severely impacts travellers' travel time. In this paper, we investigate the use of congestion prediction in velocity forecasting models. Using a data-driven approach, we classify traffic observations into classes with and without congestion. We find that this classification improves velocity forecasting, showing that using congestion as an attribute reduces the MAE by at least 6.15% for different machine and deep learning models including random forests, multi-layer perceptrons and recurrent neural networks. We propose a random forest model that identifies the future congestion state from past traffic velocity and volume data and then use it to build new short-term velocity forecasting models. These models reduce the MAE prediction error up to 3.37% over the best models that do not consider congestion. This improvement represents overcoming a 53.75% of the error due to not precisely knowing the future congestion state. [ABSTRACT FROM AUTHOR]

Errors in Cyber-Physical Systems present a major problem given the current state of technological complexity. Self-diagnosis can contribute to address it, being the standpoint of the present paper. Hence, an application of exploratory Machine Learning models to assess the functioning of robot software in order to identify anomalies that lead to low performance is proposed. More precisely, Hybrid Unsupervised Exploratory Plots (HUEPs) are extended through density-based clustering techniques, that are applied together with unsupervised exploratory projection models. As a result, intuitive and informative visualizations of software performance are obtained, supporting the monitoring and anomaly detection tasks. The proposed clustering extension of HUEPs is thoroughly validated on a massive and up-to-date open dataset, obtaining promising results. [ABSTRACT FROM AUTHOR]

This paper presents an ontology for AFDD (Automated Fault Detection and Diagnostics) of HVAC (Heating, Ventilation, and Air conditioning) systems in buildings called "AFDDOnto". Presently, the AFDD models are mainly data-centric and often lack semantic information such as contextual information and spatial information; additionally, configuration information, analysis, and results used for model development are lost once developed. This impedes an effective mechanism for tracking changes and updating the model for future developments and use cases. The Proposed ontology can be used for AFDD model development, tracking changes, analytics, visualization, and digital twinning by enabling integration of BIM with BAS/BMS (Building Automation System/Building Management System) concepts and secondly to store AFDD configuration and analytics in the AFDDOnto. Select competency questions are constructed using SPARQL queries to access the proposed knowledge model. The proposed ontology has been tested against different measures using multiple metrics and a case study and further validated using a semi-structured survey of experts. Applied AI engineers, Facility managers, Asset managers, and building owners aiming to develop AFDD models for HVAC systems can benefit from adopting this ontology for HVAC maintenance, including analysis, model development, and knowledge management. [ABSTRACT FROM AUTHOR]

Seismic intensity measures (IMs) quantify the severity of ground motions and their impacts on structures. They play a vital role in many aspects of earthquake engineering. This paper proposes a novel method, namely the express iteration method (EIM), for constructing effective vector-valued IMs based on dozens of existing scalar ones given a specific engineering structure or a class of them. Taking advantage of the sophisticated while efficient mapping between scalar IMs and engineering demand parameters (EDPs) via a machine learning model, EIM iteratively eliminates less important scalar IMs from a pool of candidates to find the most effective combinations for a vector-valued IM and achieves superior computational efficiency by avoiding updating the nonlinear mapping during the process. Taking a base-isolated structure and its non-isolated counterpart for a demonstrating case study, the performance of the vector-valued IMs determined by EIM is compared with those by other existing methods in the literature for the task of selecting the most unfavorable ground motions. The results show that EIM prioritizes records with the largest peak inter-story drift PIDs and thus leads to the smallest subset that imposes most severe structural damage, while its computational cost was two orders of magnitude smaller as compared to the existing methods of similar effectiveness. Such superior performance can also be expected in all tasks that involve vector-valued IMs, including but not limited to multi-dimensional fragility analysis, incremental dynamic analysis, and real-time seismic damage prediction. [ABSTRACT FROM AUTHOR]

This paper studies the approximate optimal event-triggered tracking control problem for a class of second-order nonlinear systems with prescribed performances. By employing a prescribed performances function, restrictions on tracking errors are removed for prescribed performances including convergence rate, maximum overshoot, and steady-state error. On this basis, a complex optimal tracking control problem can be transformed into an optimal stabilising control problem by constructing an augmented system. Subsequently, an event-triggered mechanism is introduced to reduce communication resources. Then, an adaptive critic learning algorithm is used to solve the Hamilton-Jacobi-Bellman equation, where the weights in the critic networks are tuned through the gradient descent approach and experience replay technology. Through the developed learning method, the persistence of excitation condition is released, and the data efficiency is improved. The Lyapunov stability theory is introduced to verify the stability of the close-loop system and the tracking errors are uniformly ultimately bounded. Finally, simulation results are presented to show the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Crop classification is important for monitoring crop growth and ensuring national food security. Crop plots have a complex spatial planting structure with a certain degree of fragmentation, which leads to mixing of different crops with unclear borders, in turn affecting the ability of the classification model to recognize and distinguish between crops. To address the above problems, this paper proposes a crop classification method based on multi-source optical remote sensing data, which can extract crop information more accurately by taking advantage of the multi-spectral features and different spatial resolutions of different optical data. Firstly, spectral bands and vegetation indexes are extracted from Sentinel-2B and Jilin Gaofen (JLGF02B). Then, a Gemel Convolutional Neural Network (G-CNN) is constructed to fully mine and integrate the feature information at different scales for crop classification. Finally, the method is compared with classic deep learning algorithms. The result shows that the proposed G-CNN algorithm gets the best results with an overall accuracy OA of 96.5% and a Kappa coefficient of 94.7%. The G-CNN model provides a new idea and method for crop classification using multi-scale optical data. [ABSTRACT FROM AUTHOR]

AbstractAdvancing sustainable efforts in hotels, especially small and medium-sized enterprises (SMEs), is important for addressing environmental and social impacts and promoting long-term viability in the hospitality industry. This paper uses machine learning techniques to classify sustainable practices in SMEs. Analyzing data from Booking.com’s “Travel Sustainable Level” section, which includes 30 sustainability measures across five categories (waste, water, energy and greenhouse gases, nature, and destination and community), this study examines 19 factors affecting hotel sustainability. Non-linear machine learning models identify significant features: performance, star rating, size, group membership, and coastal location. These findings are validated through a model introducing ‘Levels of Commitment to Sustainability, “categorizing hotels based on their adherence to these measures”. Variables were more distinctly differentiated by commitment levels than Booking.com levels. Post-hoc analysis and expert interviews reveal insights into the least adopted sustainability measures and their perceived costs and benefits. This study introduces a novel classification approach for hotel sustainability, providing essential insights for effective sustainable hotel management. It highlights key factors influencing sustainability and emphasizes the significance of these factors for developing targeted strategies. Furthermore, this research contributes to a broader understanding of sustainability in hospitality, demonstrating the applicability of machine learning in evaluating sustainable practices. [ABSTRACT FROM AUTHOR]

AbstractMicrobes have been extensively utilized for their sustainable and scalable properties in synthesizing desired bio-products. However, insufficient knowledge about intracellular metabolism has impeded further microbial applications. The genome-scale metabolic models (GEMs) play a pivotal role in facilitating a global understanding of cellular metabolic mechanisms. These models enable rational modification by exploring metabolic pathways and predicting potential targets in microorganisms, enabling precise cell regulation without experimental costs. Nonetheless, simplified GEM only considers genome information and network stoichiometry while neglecting other important bio-information, such as enzyme functions, thermodynamic properties, and kinetic parameters. Consequently, uncertainties persist particularly when predicting microbial behaviors in complex and fluctuant systems. The advent of the omics era with its massive quantification of genes, proteins, and metabolites under various conditions has led to the flourishing of multi-constrained models and updated algorithms with improved predicting power and broadened dimension. Meanwhile, machine learning (ML) has demonstrated exceptional analytical and predictive capacities when applied to training sets of biological big data. Incorporating the discriminant strength of ML with GEM facilitates mechanistic modeling efficiency and improves predictive accuracy. This paper provides an overview of research innovations in the GEM, including multi-constrained modeling, analytical approaches, and the latest applications of ML, which may contribute comprehensive knowledge toward genetic refinement, strain development, and yield enhancement for a broad range of biomolecules. [ABSTRACT FROM AUTHOR]

Cognitive scientists and neuroscientists typically understand the brain as a complex communication/information-processing system. A limitation of this framework is that it requires cognitive systems to have prior knowledge about their environment to successfully perform some of their basic functions, such as perceiving. It is unclear how the source of such knowledge can be explained from within this framework. Drawing on Dennett (1981), we refer to this as the loans of intelligence problem. Recent advances in machine learning have resulted in the development of a family of algorithms, including the class known as autoencoders, that seem to provide a way for the information-processing framework to avoid this problem: cognitive systems do not require loans of intelligence, but instead acquire the knowledge necessary for perception through a process of unsupervised learning. This paper argues that although autoencoders do avoid the loans of intelligence problem, how they do so should not be understood from within the information-processing framework. Instead, their success should be interpreted as a proof of concept of how neural networks can attune to Gibsonian information. We thus propose that autoencoders belong to a class of algorithms for modeling the brain that have recently been dubbed direct fit algorithms. [ABSTRACT FROM AUTHOR]

The method for identifying train axle fatigue cracks based on acoustic emission technology requires the classification and processing of acoustic emission signals. However, existing methods for recognizing axle fatigue crack acoustic emission signals, including parameter analysis, wavelet analysis, and traditional machine learning algorithms, usually rely on expert experience, resulting in low detection efficiency. Additionally, the traditional machine learning algorithms above typically have shallow structures, which limits their ability to extract deeper feature information. Therefore, this paper proposes a method that uses a one-dimensional convolutional neural network (1D-CNN) to identify train axle fatigue crack acoustic emission signals. This method achieves end-to-end intelligent recognition of acoustic emission signals, eliminating the need for manual intervention or pre-set feature extractors. Furthermore, it leverages deep learning mechanisms to deeply mine the complex features within the signals, thereby enhancing the accuracy of the model’s identification. Experimental validation was conducted using the acoustic emission signals collected from the axle fatigue crack test rig. The results demonstrate that this model can accurately identify acoustic emission signals of axle fatigue cracks, achieving an identification accuracy of over 99%. Meanwhile, compared with other network models, the effectiveness of the proposed model is further verified. [ABSTRACT FROM AUTHOR]

Studies of online campaigning have consistently demonstrated a positive impact on electoral success, but it remains unclear how this occurs. Some find the content and style of post matter, while others have pointed to overall activity as the key driver, promoting a “broadcast” effect model. Still, others have argued for indirect effects whereby candidates rely on followers to share content within their networks. This paper develops a “joined-up” model to test these arguments that includes new measures to capture the responsiveness of candidate tweets and the extent of user engagement. We apply the model to the 2017 UK General Election Twitter campaign. Our findings confirm a digital campaign effect, but in a two-step rather than direct manner. Specifically, candidates that attract more engagement with their tweets (likes and retweets) enjoy more electoral success. We expand on our findings to argue for a “network” rather than “broadcast” model of digital campaign effects. [ABSTRACT FROM AUTHOR]

AbstractEarly detection of plum bruising is important in the online postharvest quality sorting process. This paper explores the rapid detection of bruises in plum at five stages (1, 3, 6, 24, and 48 h after bruising) using hyperspectral imaging system. Spectral preprocessing was performed using five methods (unit vector normalization, multiplicative scattering correction, standard normal variate, detrending, and baseline). The support vector machine was established to discriminate the spectral samples of healthy and bruised plums at stage 1 on full wavelengths. The results indicated that the spectral data pretreated by multiple scattering correction yielded better results. The characteristic wavelengths of the spectra were selected by six algorithms (random frog, competitive adaptive reweighted sampling, variable combination population analysis, successive projection algorithm, uninformative variable elimination and bootstrapping soft shrinkage). Subsequently, the support vector machine was established at stage 1 based on these selected characteristic wavelengths. Comparing the results of models, the support vector machine based on the characteristic wavelengths selected by competitive adaptive reweighted sampling generated a satisfied effect with an accuracy of 95% for the calibration set and 98% for the prediction set. This model was selected for bruise detection in plums at the residual stage. The characteristic wavelength grayscale images selected based on the model were used to successfully visualize the plum bruise regions in five stages using minimum noise fraction transformation and the principal component analysis method. The results showed that the hyperspectral imaging technique combined with machine learning algorithm could be used to identify plum bruises at different stages. This study contributes to the development of an online detection system for bruises in plum. [ABSTRACT FROM AUTHOR]

Over the past twenty years, higher education institutions (HEIs) including universities, polytechnics, and vocational education providers, have become major conduits for the teaching and dissemination of the sustainable development goals (SDGs). The Times Higher Education Impact rankings, other rankings schemes and external accreditation requirements also pressure HEIs to report their progress towards the SDGs periodically. For large HEIs, reporting requirements can equate to significant resource burdens, and these can be insurmountable. In this paper, we develop a novel approach to measuring SDG content in the course descriptions and learning outcomes of 5461 courses offered by a major Australian university. Our method utilises semantic matching, an AI-based technique that assesses similarities between key terms in the dataset (including course descriptions and course learning outcomes) and those in each SDG. Our results achieve a 75% fit to the data and show that Quality Education (SDG 4) and Partnerships for the Goals (SDG 17) are the most common SDGs covered at the case university and that these reflect course learning outcomes that focus on pedagogy. Where SDGs are more domain-specific, there is a much less consistent coverage across the university, and this is observable at the Faculty and School levels. This highlights that it is not possible or desirable to cover all SDG content in all educational offerings consistently. Our algorithm is a potential solution for larger HEIs that seek to capture and report the SDG contributions of their education offerings holistically using existing textual datasets, without the need for excessive resource investments. [ABSTRACT FROM AUTHOR]

Remaining Useful Life (RUL) prediction of rolling bearings is one of the intricate and important issues for equipment intelligent maintenance and health management. Various machine learning models and methods have been applied to rolling bearing RUL prediction. However, a single model cannot effectively extract state information and obtain accurate prediction results, and its generalisation is not stable under the condition of small sample data. Therefore this paper proposes an intelligent hybrid deep learning model for achieving accurate RUL prediction of rolling bearings. Firstly, the one-dimensional vibration signal is transformed into the corresponding two-dimensional time-frequency diagram via Continuous Wavelet Transform (CWT). Secondly, the diagram is input into a Multilayer Perceptron (MLP) consisting of a basic three-layer feed-forward network to obtain a one-dimensional feature vector. And lastly, the obtained feature vector is input into an integrated model based on Deep Autoregressive and Transformer to produce the probability distribution and obtain the prediction results of rolling bearing RUL. Extensive experiments on two rolling bearing datasets show that the proposed model outperforms six other comparative models in extracting bearing fault features and predicting bearing RUL, which demonstrates that the proposed model can effectively extract bearing fault features and accurately predict bearing remaining useful life. [ABSTRACT FROM AUTHOR]

The development of next-generation sequencing (NGS) opens opportunities for new applications such as liquid biopsy, in which tumor mutation genotypes can be determined by sequencing circulating tumor DNA after blood draws. However, with highly diluted samples like those obtained with liquid biopsy, NGS invariably introduces a certain level of misclassification, even with improved technology. Recently, there has been a high demand to use mutation genotypes as biomarkers for predicting prognosis and treatment selection. Many methods have also been proposed to build classifiers based on multiple loci with machine learning algorithms as biomarkers. How the higher misclassification rate introduced by liquid biopsy will affect the performance of these biomarkers has not been thoroughly investigated. In this paper, we report the results from a simulation study focused on the clinical utility of biomarkers when misclassification is present due to the current technological limit of NGS in the liquid biopsy setting. The simulation covers a range of performance profiles for current NGS platforms with different machine learning algorithms and uses actual patient genotypes. Our results show that, at the high end of the performance spectrum, the misclassification introduced by NGS had very little effect on the clinical utility of the biomarker. However, in more challenging applications with lower accuracy, misclassification could have a notable effect on clinical utility. The pattern of this effect can be complex, especially for machine learning-based classifiers. Our results show that simulation can be an effective tool for assessing different scenarios of misclassification. [ABSTRACT FROM AUTHOR]

To achieve precise prediction of the air pressure changes in the critical components of the air brake system, including brake pipes (BP), auxiliary reservoirs (AR), and brake cylinders (BC), this paper proposed a neural network-based air brake model developed from the brake application and release experimental data with different brake applications (50, 70, 100, and 150 kPa) conducted on a heavy haul train (HHT) test platform. The physical structures of the air brake system and the working mechanisms of a neural network are introduced, and the neural network-based air brake model scheme is presented by analysing the variation of air pressure within the air brake system under typical braking conditions (50 kPa). Besides, by comparing with other models, the superiority of the proposed model is verified, and the proposed model achieves higher computational efficiency, which increased by about 2900 times compared to the fluid dynamic model under the same conditions. Then, the model is applied to the calculation of the longitudinal dynamics of a 10,000-ton HHT, and the results demonstrate that the proposed model could serve as a promising tool for solving the braking excitation problem of longitudinal dynamics of trains and the study of intelligent braking control of trains. [ABSTRACT FROM AUTHOR]

Machine learning-based prediction of the output characteristics of a negative capacitance (NC) double gate (DG) Tunnel FET (TFET) (NC DG TFET) has been demonstrated in this paper. On current (ION), OFF current (IOFF) and sub-threshold slope (SS) have been predicted for different values of the body thickness (TSi), oxide thickness (TOX) and channel length (LCh) and the work function (WF). Random Forest Regression (RFR) model is used in the ML framework to explore the device characteristics. A total of 5444 simulated data has been generated by using the SILVACO ATLAS tool. After pre-processing the data, RFR-ML has been applied to obtain maximum accuracy. Root means square error (RMSE) (~0.34), R2 score (~0.965) and accuracy (~98.51%) parameter have been used as the figures of merit (FOM) to judge the efficiency of the proposed model with respect to the state-of-the-art literature. Besides, with higher accuracy, the model is capable of predicting the data within a few seconds compared to the few hundred hours using TCAD tools. This fast and accurate design strategy will surely be efficacious for the present VLSI industries. [ABSTRACT FROM AUTHOR]

This paper studies the combination of random forest (RF) and classical statistical modeling. We propose two algorithms: RF cluster + ridge and RF regression + ridge, in which the RF reduces overfitting and has a good anti-noise ability. The ridge regression can effectively handle the correlated data by the ridge penalty. To test the performance of the proposed techniques, we model and track the S&P 500 index. Index tracking aims to replicate the returns of an index fund and is very popular in finance. We compare the performance of the proposed methods with those of several existing methods. The portfolio obtained by each technique is quantified using the tracking error. We find that this combination always has higher accuracy. Empirical results show that the proposed methods are qualified for index funds. Our findings shed light on the synthesis of machine learning and statistical modeling techniques and provide an effective practice of this combination. [ABSTRACT FROM AUTHOR]

AbstractTopological indices are well-known molecular descriptors that may be applied to any graph and to model specific molecular structures, utilized for property correlation in quantitative structure activity and property relationship (QSAR and QSPR) analyses. Strong methods for machine learning information and thermodynamic simulations of massive networks and nanomaterials can be facilitated by using topological indices. Kekulenes are a novel type of cycloarenes, arranged in doughnut-shaped cyclic benzene rings with riveting superaromaticity and other significant properties. The main aim of this paper is to ascertain the notable degree and the neighborhood degree sum-based indices through M-polynomials and neighborhood M-polynomials of certain types of tessellations of kekulenes. Then, a graphical comparison of the resulting indices is also performed. [ABSTRACT FROM AUTHOR]

Clay swell potential can be classified based on the value of activity and it is defined as the ratio of plasticity index to clay content as a percentage. This paper outlines the investigation into how activity correlates with other key properties of clayey soils. Specifically, four approaches were evaluated for predicting activity using: (a) liquid limit (LL), specific surface area (SSA), cation exchange capacity (CEC) and clay content; (b) LL, SSA and CEC; (c) LL; and (d) SSA and CEC. For this purpose, a database of 104 samples was collected from which 35 machine learning algorithms were trained. Gradient Boosting Trees showed the highest prediction accuracy in the four approaches and, to improve its prediction performance, a Bayesian Optimization was applied to tune theirs hyperparameters, resulting in the final models. The performance of the developed models was evaluated, showing prominent results with exceptionally good metrics, except in the approach from SSA and CEC where the trained algorithm was not capable of predicting activity with confidence (R2=0.46). This algorithm can predict activity using only LL with high accuracy (R2=0.94), and when combined with SSA and CEC, the precision is further enhanced (R2=0.96). [ABSTRACT FROM AUTHOR]