Standard Machine Research Articles

Machine embroidery of light-emitting textiles with multicolor electroluminescent threads.

Advances in electroluminescent threads, suitable for weaving or knitting, have opened doors for the development of light-emitting textiles, driving growth in the market for flexible and wearable displays. Although direct embroidery of these textiles with custom designs and patterns could offer substantial benefits, the rigorous demands of machine embroidery challenge the integrity of these threads. Here, we present embroiderable multicolor electroluminescent threads-in blue, green, and yellow-that are compatible with standard embroidery machines. These threads can be used to stitch decorative designs onto various consumer fabrics without compromising their wear resistance or light-emitting capabilities. Demonstrations include illuminating specific messages or designs on consumer products and delivering emergency alerts on helmet liners for physical hazards. Our research delivers a comprehensive toolkit for integrating light-emitting textiles into trendy, customized crafts tailored to the unique requirements of diverse flexible and wearable displays.

A Comparison of Standard Statistical, Machine Learning and Deep Learning Methods in Forecasting the Time Series

A Comparison of Standard Statistical, Machine Learning and Deep Learning Methods in Forecasting the Time Series

Increasing efficiency of induction motor by predictive control system

Standard squirrel-cage induction machine is widely used in industrial enterprises due to its advantages in terms of reliability, durability and economic efficiency. However, one of its disadvantages is that it suffers from a power dissipation factor due to its inability to balance active and reactive power consumption. Especially when the SCIM is running at no load condition or during the initial start-up, the power factor or useful operation and performance are drastically reduced. Therefore, improving the power factor of the induction machine is an actual topic for the following article, and has had some interesting solutions for several years. In this thesis, two different induction motor control algorithms (frequency converter and Predictive torque control) are included and a clear comparison between them is given. The main focus of this article is to design an induction motor control system using the two algorithms mentioned above, analyse the performance of different control methods, and experimentally verify these algorithms by comparing simulations and experimental results.

Spiking neural network with local plasticity and sparse connectivity for audio classification

Purpose. Studying the possibility of implementing a data classification method based on a spiking neural network, which has a low number of connections and is trained based on local plasticity rules, such as Spike-Timing-Dependent Plasticity. Methods. As the basic architecture of a spiking neural network we use a network included an input layer and layers of excitatory and inhibitory spiking neurons (Leaky Integrate and Fire). Various options for organizing connections in the selected neural network are explored. We have proposed a method for organizing connectivity between layers of neurons, in which synaptic connections are formed with a certain probability, calculated on the basis of the spatial arrangement of neurons in the layers. In this case, a limited area of connectivity leads to a higher sparseness of connections in the overall network. We use frequency-based coding of data into spike trains, and logistic regression is used for decoding. Results. As a result, based on the proposed method of organizing connections, a set of spiking neural network architectures with different connectivity coefficients for different layers of the original network was implemented. A study of the resulting spiking network architectures was carried out using the Free Spoken Digits dataset, consisting of 3000 audio recordings corresponding to 10 classes of digits from 0 to 9. Conclusion. It is shown that the proposed method of organizing connections for the selected spiking neural network allows reducing the number of connections by up to 60% compared to a fully connected architecture. At the same time, the accuracy of solving the classification problem does not deteriorate and is 0.92...0.95 according to the F1 metric. This matches the accuracy of standard support vector machine, k-nearest neighbor, and random forest classifiers. The source code for this article is publicly available: https://github.com/sag111/Sparse-WTA-SNN.

A special face mask with integrated angle meter to optimize the head position for bag-valve-mask ventilation—a prospective simulated proof-of-concept study

The supine position in unconscious patients resulted in the tongue falling back because of the loss of muscle tension which obstructed the upper airway. The head reclination opened the airway, however unexperienced rescuers faced difficulties in performing bag-valve-mask ventilation as it was hard to detect the optimal head position for opening the airway. The bag-valve-mask ventilation could be optimized if it indicated an optimized head angle to the rescuer. A digital sensor was attached to the face mask for measuring the degree of head reclination. Moreover, the face mask was equipped with an airway trainer and sealed via adhesive tape. The airway trainer was connected to a test lung and ventilated through pressure-controlled standard anesthesia machine (Pmax = 10 mbar, Positive End Expiratory Pressure (PEEP) = 0 mbar, f = 12/min). The head of airway manikin was tilted to 42 degrees starting from neutral position with a gradient of 2 degrees. The primary endpoint was the correlation of preset angles and digitally measured head position angles via the mask. In addition, the tidal and minute volumes were evaluated based on the head reclination. The preset and the digitally measured head position angles depicted correlation (R2 = 0.9895855684; p < 0.001). The tidal volume was 150 mL in head position angle of <10 degrees, 200 mL at 18 degrees, 450 mL at 25 degrees, and it levelled with about 500 mL at 30 degrees. In conclusion, digital head position angle was measured to detect the correct head position. A signal in the face mask could indicate the optimized head position to the first respondents or inexperienced rescuers in emergency ventilation.

Genetic Artificial Hummingbird Algorithm-Support Vector Machine for Timely Power Theft Detection.

Utilities face serious obstacles from power theft, which calls for creative ways to maintain income and improve operational effectiveness. This study presents a novel hybrid genetic artificial hummingbird algorithm-support vector machine classifier to detect power theft. The proposed algorithm combines the artificial hummingbird algorithm exploration phase with the genetic algorithm's mutation and crossover operators, to optimize the support vector machine's hyperparameters and categorize users as fraudulent or nonfraudulent. It utilizes 7,270 rows of labeled historical electricity consumption data from the Liberia Electricity Corporation over 15 independent runs. The methodology entailed data preprocessing, data split into training, validation, and testing sets in an 80-10-10 ratio, z-score normalization, optimization, training, validation, testing, and computation of six evaluation metrics. Its performance is compared with 13 metaheuristic classifiers and the conventional support vector machine. Findings indicate that the genetic artificial hummingbird algorithm-support vector machine outperforms the 13 rivals and the standard support vector machine in the six assessment measures with an accuracy score of 0.9986, precision of 0.9971, f_score of 0.9986, recall of 1, Matthews correlation coefficient of 0.9972, and g_mean of 0.9987. Furthermore, 90% of the time, Wilcoxon rank-sum tests revealed statistically significant differences between the algorithm and its rivals, demonstrating its superiority. The average run time is 4,656 seconds, the 3rd highest among its competitors. Despite the time complexity trade-off, its excellent performance on the unimodal and multimodal benchmark test functions, placing joint best in 7 out of 7 and 5 out of 6, respectively, provides important insights into the model's capacity to balance exploitation and exploration, improve local search, and avoid becoming stuck in the local optimum. These findings address important metaheuristic optimization gaps highlighting the model's potential for power theft detection.

Behavioral marker-based predictive modeling of functional status for older adults with subjective cognitive decline and mild cognitive impairment: Study protocol.

This study describes a research protocol for a behavioral marker-based predictive model that examines the functional status of older adults with subjective cognitive decline and mild cognitive impairment. A total of 130 older adults aged ≥65 years with subjective cognitive decline or mild cognitive impairment will be recruited from the Dementia Relief Centers or the Community Service Centers. Data on behavioral and psychosocial markers (e.g. physical activity, mobility, sleep/wake patterns, social interaction, and mild behavioral impairment) will be collected using passive wearable actigraphy, in-person questionnaires, and smartphone-based ecological momentary assessments. Two follow-up assessments will be performed at 12 and 24 months after baseline. Mixed-effect machine learning models: MErf, MEgbm, MEmod, and MEctree, and standard machine learning models without random effects [random forest, gradient boosting machine] will be employed in our analyses to predict functional status over time. The results of this study will be fundamental for developing tailored digital interventions that apply deep learning techniques to behavioral data to predict, identify, and aid in the management of functional decline in older adults with subjective cognitive decline and mild cognitive impairment. These older adults are considered the optimal target population for preventive interventions and will benefit from such tailored strategies. Our study will contribute to the development of self-care interventions that utilize behavioral data and machine learning techniques to provide automated analyses of the functional decline of older adults who are at risk for dementia.

Hybrid prognosis of drill-bits based on direct inspection

Within the paradigm of industry 5.0, manufacturing systems are seeking for human-centred production, where the operator finds high-level supervision tasks. In this context, low-level decision making should be performed by machines themselves. In this paper, a hybrid prognosis algorithm is developed to automatically inspect the cutting edges of drill-bits and to predict their Remaining Useful Life (RUL) and the associated probability density function. The solution relies on the automatic measurement of flank wear through convolutional filtering and edge detection. Prognosis exploits particle filter, which updates multi-layer perceptron with online data, to adaptively predict drill-bits RUL. The solution reduces the experimental preliminary run-to-failures needed for training standard machine learning algorithms, exploiting them in a real-time adaptive scenario, while predicting tool RUL under untested and variable cutting process operations. The algorithm uses direct wear observations, taken during set-up times (e.g., tool changes, workpiece change), thus not interfering with the process.

LERCause: Deep learning approaches for causal sentence identification from nuclear safety reports.

Identifying causal sentences from nuclear incident reports is essential for advancing nuclear safety research and applications. Nonetheless, accurately locating and labeling causal sentences in text data is challenging, and might benefit from the usage of automated techniques. In this paper, we introduce LERCause, a labeled dataset combined with labeling methods meant to serve as a foundation for the classification of causal sentences in the domain of nuclear safety. We used three BERT models (BERT, BioBERT, and SciBERT) to 10,608 annotated sentences from the Licensee Event Report (LER) corpus for predicting sentence labels (Causal vs. non-Causal). We also used a keyword-based heuristic strategy, three standard machine learning methods (Logistic Regression, Gradient Boosting, and Support Vector Machine), and a deep learning approach (Convolutional Neural Network; CNN) for comparison. We found that the BERT-centric models outperformed all other tested models in terms of all evaluation metrics (accuracy, precision, recall, and F1 score). BioBERT resulted in the highest overall F1 score of 94.49% from the ten-fold cross-validation. Our dataset and coding framework can provide a robust baseline for assessing and comparing new causal sentences extraction techniques. As far as we know, our research breaks new ground by leveraging BERT-centric models for causal sentence classification in the nuclear safety domain and by openly distributing labeled data and code to enable reproducibility in subsequent research.

Machine learning classification based on k-Nearest Neighbors for PolSAR data.

In this work, we focus on obtaining insights of the performances of some well-known machine learning image classification techniques (k-NN, Support Vector Machine, randomized decision tree and one based on stochastic distances) for PolSAR (Polarimetric Synthetic Aperture Radar) imagery. We test the classifiers methods on a set of actual PolSAR data and provide some conclusions. The aim of this work is to show that suitable adapted standard machine learning methods offer excellent performances vs.computational complexity trade-off for PolSAR image classification. In this work, we evaluate well-known machine learning techniques for PolSAR (Polarimetric Synthetic Aperture Radar) image classification, including K-Nearest Neighbors (KNN), Support Vector Machine (SVM), randomized decision tree, and a method based on the Kullback-Leibler stochastic distance. Our experiments with real PolSAR data show that standard machine learning methods, when adapted appropriately, offer a favourable trade-off between performance and computational complexity. The KNN and SVM perform poorly on these data, likely due to their failure to account for the inherent speckle presence and properties of the studied reliefs. Overall, our findings highlight the potential of the Kullback-Leibler stochastic distance method for PolSAR image classification.

Harnessing Twitter: Sentiment Analysis for Predicting Election Outcomes in India

This paper explores the potential of Twitter, a popular social media platform, as a tool for predicting election outcomes. Sentiment analysis has emerged as a powerful tool for predicting election outcomes, with numerous studies showcasing its effectiveness in various countries. For instance, research has utilized sentiment analysis to forecast election results in nations like the USA, India, Pakistan and other countries, demonstrating the utility of social media data in gauging public opinion and predicting electoral results [1]. Elections in India are always considered important events that most people look forward to the rapid growth of social media in the past has provided end users with powerful tools to share their ideas. Twitter, which is one such platform, provides daily updates on political events through various hashtags and trends. People react to political events and give their opinions. Our approach is to collect tweets from top political parties contesting the Gujarat Assembly Elections 2022, and then calculate sentiment scores. The database includes a variety of recent and well-liked tweets about a specific political party. Party tweets are generated with specific keywords like “BJP”, “AAP”,“Congress” and so on. In the context of India, Twitter sentiment analysis tools and classification have been used to predict the outcomes of state assembly elections, underscoring the potential of social media data in forecasting electoral results within the country [2]. We used standard machine learning algorithms like VADER sentiment analyzer on Random Forest and Decision Tree for our classification and testing data to classify tweets as positive and negative. As a result, this work uses sentiment analysis to evaluate tweets gathered from Twitter and forecast election outcomes. This work shows the growing influence of social media on politics and the feasibility of using such platforms for predictive analysis. The findings of this study could provide valuable insights for political parties, policymakers, and researchers interested in the intersection of social media and politics. Random Forest and Decision Tree models performed well in predicting election outcomes based on sentiment analysis on Twitter data with 89% and 86% respectively.

Estimating R, L and C in Arithmetic Circuits using ML

In Very Large Scale Integration (VLSI) circuits, the estimation techniques for resistors (R), inductors (L), and capacitors (C) heavily rely on segmented circuit analysis, which involves usage of complex mathematical simplification models. These methods have been conventionally applied to estimate the behavior of circuits, but when faced with systems featuring unique circuit architectures, they often encounter inaccuracies and limitations. The significance of adders as fundamental building blocks in intricate circuit design cannot be overstated. In such complex circuits, various parameters, including parasitic resistances, inductances, and capacitances, engage an indispensable effort in the analysis of delays and performance. However, the conventional estimation methods fail to address the complexities of these systems and the impacts of parasitics accurately. To overcome these challenges, this paper proposes a novel approach that harnesses the potential of machine learning algorithms. By integrating machine learning into the estimation process, the research aims to achieve specific and precise analysis methods that can cater to the needs of modern VLSI circuits. The proposed methodology involves collecting a comprehensive dataset using different adder circuits. From each individual adder circuit layout, the relevant information on resistors, capacitors, and inductors is carefully extracted and compiled. This dataset serves as the foundation for training the machine learning models. Three standard machine learning models are employed in this study: adaboost, Tree, and k-Nearest Neighbors (kNN). Their task is to predict the values of resistors, inductors, and capacitors based on the input data from the adder circuits. Among these models, adaboost proves to be the most effective, exhibiting superior performance by achieving a reduced root mean square error of about 0.008. When compared to the Tree and kNN models, adaboost stands out as the optimal choice for accurately estimating the R values in VLSI circuits.

S-TBN: A New Neural Decoding Model to Identify Stimulus Categories From Brain Activity Patterns.

Neural decoding is still a challenging and a hot topic in neurocomputing science. Recently, many studies have shown that brain network patterns containing rich spatiotemporal structural information represent the brain's activation information under external stimuli. In the traditional method, brain network features are directly obtained using the standard machine learning method and provide to a classifier, subsequently decoding external stimuli. However, this method cannot effectively extract the multidimensional structural information hidden in the brain network. Furthermore, studies on tensors have show that the tensor decomposition model can fully mine unique spatiotemporal structural characteristics of a spatiotemporal structure in data with a multidimensional structure. This research proposed a stimulus-constrained Tensor Brain Network (s-TBN) model that involves the tensor decomposition and stimulus category-constraint information. The model was verified on real neuroimaging data obtained via magnetoencephalograph and functional mangetic resonance imaging). Experimental results show that the s-TBN model achieve accuracy matrices of greater than 11.06% and 18.46% on the accuracy matrix compared with other methods on two modal datasets. These results prove the superiority of extracting discriminative characteristics using the STN model, especially for decoding object stimuli with semantic information.

A Deep Learning Approach to Real-time Electricity Load Forecasting

In light of the increasing importance of accurate and real-time electrical demand forecasting, this research presents a deep learning model with the goal of dramatically improving predictive accuracy. Conventional methods of forecasting, such as linear regression, have trouble capturing the complex patterns included in data about electricity usage. Standard machine learning methods are shown to be wanting when compared to the suggested deep Long Short-Term Memory (LSTM) model. Mean Absolute Error (MAE) of 5.454 and Mean Squared Error (MSE) of 18.243 demonstrate the deep LSTM model's proficiency in tackling this problem. The linear regression, on the other hand, achieved a MAE of 47.352 and an MSE of 65.606, which is lower than the proposed model. Because of its greater predictive precision and reliability, the deep LSTM model is a viable option for accurate, real-time prediction of electricity demand.

Dynamic and Steady State Modelling of the Doubly Fed Twin Stator Induction Generator With Core Loss.

This paper presents dynamic and steady state models of the brushless doubly fed twin stator induction machine (BDFTSIM) with emphasis on its use as a generator. The dynamic models are presented in the general reference frame. The models include the effects of core loss on both the stator and the rotor. The frequency on the rotor of the BDFTSIM is much higher than in a standard induction machine so that core losses can not be neglected. The performance analysis indicates the feasibility of the BDFTSIM as a variable speed wind generator

An efficient wavelet thresholding strategy and robust shrinkage approach for de‐noising ECG signal

AbstractThis paper introduces an efficient wavelet thresholding strategy and robust shrinkage approach (WTS) for de‐noising the ECG signals. The optimal mother wavelet for de‐noising the ECG signal is automatically selected based on cross‐correlation and energy‐to‐entropy indices. A dynamic threshold is applied to various levels of decomposition to eliminate different types of noise. A shrinkage approach is suggested to attain seamless on–off transitions. The achievement of the WTS is compared with the state of the art reported in the literature using 180 real ECG signals collected from 60 volunteers. The ECG signals were recorded using three standard ECG machines without applying the filtering option to provide a diverse range of noisy ECG signals. The de‐noising outcomes show that the presented WTS can automatically select the optimal mother wavelet for the particular ECG signal, and it is superior to other techniques in terms of the average predicted signal‐to‐noise ratio (), the average predicted root‐mean‐square error (), and the average execution time (AET).

Panel data nowcasting: The case of price–earnings ratios

AbstractThe paper uses structured machine learning regressions for nowcasting with panel data consisting of series sampled at different frequencies. Motivated by the problem of predicting corporate earnings for a large cross‐section of firms with macroeconomic, financial, and news time series sampled at different frequencies, we focus on the sparse‐group LASSO regularization which can take advantage of the mixed‐frequency time series panel data structures. Our empirical results show the superior performance of our machine learning panel data regression models over analysts' predictions, forecast combinations, firm‐specific time series regression models, and standard machine learning methods.

On learning rates for regularized Nyström subsampling in unsupervised domain adaptation

The regularized Nyström subsampling is a popular approach for learning problems that deals with big data. We employ such technique in the context of the unsupervised domain adaptation problems with covariate shift assumption. Within the framework of the Reproducing Kernel Hilbert Space concept, an algorithm is constructed that is a combination of the Nyström subsampling and the two-steps iterated Tikhonov regularization. This approach allows significantly reduce the amount of computing resources involved and at the same time maintains the same learning rates as for the standard machine learning algorithms.

Identifying Thyroid Dysfunction Using Standard Laboratory Testings – A Systematic Review

Thyroid dysfunction includes various thyroid-related illnesses, with subclinical hypothyroidism or hyperthyroidism at the initial stage. Blood TSH and T4 levels indicate a neutral ground between clinical conditions. Following guidelines for levothyroxine administration based on thyroid hormonal levels is beneficial for hypothyroid patients. TSH, a measurable signal, is crucial for assessing thyroid activity, with reference ranges determined by testing facilities. However, elevated TSH levels require consideration of the patient's history and lifestyle before intervention. A Machine Learning-based Thyroid Dysfunction Identification (ML-TDI) model was developed to screen individuals for medicinal intervention. Despite its prevalence and health consequences, thyroid dysfunction often goes undiagnosed. The study used standard laboratory data and machine learning algorithms to identify thyroid dysfunction, suggesting the potential for technology-driven screenings during routine medical procedures. However, assessing the benefits and risks of widespread thyroid illness evaluation requires well-conducted randomized trials.

Development of a Simulator for Household Refrigerator Using Equation-Based Optimization Control with Bayesian Calibration

Conventional household refrigerators consist of a motor-driven compressor, evaporator, condenser, and expansion valve. To determine the optimal operation strategies of refrigerators, it is essential to investigate the overall system performance, using an appropriate simulator. This study proposed a data-driven simulator based on engineering features and machine learning algorithms for conventional household refrigerators. The most correlated variables for identifying the indoor temperature of refrigerators were identified using variable importance, and these were revealed to be the circulation fan speed, compressor operation status, and refrigerant flow direction. A data-driven simulator was constructed using Bayesian calibration, which considers the important variables, combined with a straightforward heat balance equation. The Markov Chain Monte Carlo approach was used to simultaneously calibrate three coefficients on the critical variables based on the heat balancing equation on each time step, which is consistent with the actual temperature of the container. The results revealed that the proposed approach (equation-based Bayesian calibration outperforms) standard machine learning algorithms, such as linear regression and random forest models, by 38.5%. Additionally, compared to the typical numerical analysis method, it can reduce the delivery time and effort required to develop a reliable simulator for household refrigerators.