Search

One of the main concerns of cloud-based services based on machine and deep learning algorithms is the privacy of users' data. This is particularly relevant when companies want to leverage such services because they have to outsource potentially sensible data to be processed. In this work, the problem of privacy-preserving anomaly detection on industrial vibrational data with machine learning is tackled. It consists in the detection of irregularities or deviations from expected patterns in the vibration signals generated by industrial machinery and equipment. Such anomalies can be indicative of potential equipment failures, maintenance needs, or process deviations, making their timely detection critical for ensuring the smooth operation and reliability of industrial systems. We combine this industrial need with the ability to guarantee data privacy by proposing encrypted vibrational anomaly detection (EVAD). EVAD allows the detection of anomalies on vibrational data in a privacy-preserving manner by integrating, for the first time in the literature, one-class support vector machines and homomorphic encryption, the latter being a particular kind of encryption that allows the computation of some operations directly on encrypted data. Experimental results show that, on two publicly available datasets for vibrational anomaly detection, EVAD is able to distinguish, in a privacy-preserving manner, between nominal and anomaly situations, in an effective and efficient way. To the best of our knowledge, EVAD represents the first privacy-preserving solution for the detection of anomalies in vibrational data present in the literature. [ABSTRACT FROM AUTHOR]

Text classification has critical applications, including healthcare, where it can assist patients in locating specialized clinics based on their symptom descriptions. This can enhance healthcare services by reducing misdiagnoses and efficiently guiding patients to appropriate specialists. This research focuses on building a model for multi-class text classification in Arabic healthcare. Two evaluation schemes were employed using health data from the Altibbi dataset. Various feature extraction techniques were employed, including term frequency-inverse document frequency (TF-IDF) and word-to-vector (Word2Vec). Classical machine learning models such as Logistic Regression (LR), Random Forest (RF), Multinomial Naïve Bayes (MNB), Stochastic Gradient Descent, and Support Vector Machine (SVM), along with ensemble classifications, were implemented. Evaluation measures were used to compare model performance, including precision, recall, and F1-score. Results indicated that TF-IDF generally outperformed Word2Vec regarding the F1-score across most classes. LR and RF consistently demonstrated strong performance, while MNB exhibited lower precision. Ensemble methods like stacking and bagging showed promise, with LR and RF performing well across multiple classes. The SVM model displayed lower precision in certain classes. Preliminary experiments with the transformer-based models for Arabic language AraBERT and MarBERT as feature extraction and classification models demonstrated impressive precision, recall, and F1 scores across various classes. The findings show the effectiveness of ensemble models using TF-IDF and Arabic transformer models in accurately classifying Arabic health texts. However, challenges related to standardization, semantic variations, data availability, and resources should be considered. [ABSTRACT FROM AUTHOR]

This study investigates the effectiveness of six prominent machine learning models—random forest, decision trees, K-nearest neighbor, logistic regression, support vector machines, and Naïve Bayes—for intrusion detection systems in industrial Internet of Things environments. The evaluation encompasses the effects of data preprocessing techniques, including feature engineering, data normalization, recoding, and missing data mitigation. Furthermore, the research delves into dataset balancing, examining the effects of six different techniques on model performance. The investigations are conducted using the domain-specific WUSTL-IIOT-2021 dataset, which captures the unique characteristics of IIoT data. The study also investigates multi-class attack identification utilizing an innovative SMOTE-based multi-class balancing approach to tackle dataset imbalances. The results indicate that data preprocessing and intelligent dataset balancing produce consistent enhancements in the classification performance of the selected models across binary and multi-classification tasks. Random forest emerges as the standout algorithm, delivering consistently high performance with computational efficiency. [ABSTRACT FROM AUTHOR]

The catastrophic forgetting problem is one of the hotspots in the field of deep learning. At present, there is no doubt that storing samples of previous tasks in fixed-size memory is the best way to solve this problem. However, the number of samples stored in fixed-size memory is limited. With the increase in tasks, the number of samples stored in memory for a single task will decrease sharply. It is also difficult to balance the memory capacity and number of samples. To solve this problem, some methods use a fixed size of memory to store dimensionality reduction images. However, this will create new problems. 1) The quality of dimensionality reduction images is poor, and they are significantly different from original images. 2) How to choose the dimensionality reduction method of images. To address these problem, we put forward a new method. Firstly, we employ a simple and reliable scheme to solve the domain difference between dimensionality reduction images and original images. And we theoretically analyzed which image dimensionality reduction method is better. Secondly, to increase the generalization ability of our method and further mitigate the catastrophic forgetting phenomenon, we utilize a self-supervised image augmentation method and the output features similarity loss. Thirdly, we make use of the neural kernel mapping support vector machine theory to improve the interpretability of our method. Experimental results demonstrated that the top-1 average accuracy of our method is much higher than other methods when using the same size of memory. [ABSTRACT FROM AUTHOR]

This paper presents an intelligent healthcare framework by incorporating modern computing technologies like machine learning and deep learning. The sole motivation of this paper is to predict diabetes status intelligently so that patients can be made aware of their diabetes status regularly. A novel ensemble machine learning-based diabetes prediction framework, namely "iDP", has been proposed in this paper. Six different machine learning approaches have been used for the iDP framework, namely random forest, decision tree, neural network, AdaBoost, support vector machine, and XGBoost. Training and testing of ensemble modeling have been performed based on these learning techniques. A comprehensive preliminary data analysis and screening of statistical analysis have been systematically demonstrated in this paper. The recursive feature elimination and fivefold cross-validation methods have been used to select the best model. All experiments have been performed using R-Studio 1.2.5 with the help of the Rattle library 5.4.0. The preliminary data analytic and Pearson correlation coefficient values have been computed to verify the linearity relationship among the data. Results have been reported in eight performance metrics, including accuracy, specificity, sensitivity, area under curve, area under convex hull, minimum error rate, minimum weighted loss, and precision. The proposed iDP framework improves the results on different aspects in order to compare them with individual learning approaches. Despite this, a comparative performance analysis of the iDP framework has also been performed with state-of-the-art prevailing models. As per experiment, the iDP framework gives better results in most cases, like accuracy: 95.26%, AUC: 91.15%, sensitivity: 96.81%, specificity: 97.72% and precision: 90.72%. [ABSTRACT FROM AUTHOR]

Makam is defined as melodies that are described with typical the agâz(beginning), seyir (the orientation style), and karar (ending) features in a certain perde düzen (tone/fret tunings). For this reason, determining the makam is the basic step in understanding the melodic progression, as in musical keys. Machine learning, an artificial intelligence discipline, offers stable solutions for estimating different situations based on known data. In this study, classifications for makam recognition in Turkish folk music were carried out using different machine learning methods on the original dataset. The success of the different methods used was compared, and also data augmentation was performed by using SMOTE, KMeansSMOTE, ADASYN, and SVMSMOTE oversampling techniques, which are widely used in the literature to increase the prediction success of the classifiers. For each of eight different machine learning techniques, namely light gradient boosting machine, eXtreme gradient boosting, naive Bayes, decision tree, support vector machine, K-nearest neighbor, random forest, and logistic regression, in addition to classification without data augmentation, a total of 40 classification processes were carried out using four different data augmentation techniques. The results of the classification processes are reported with four different parameters: accuracy, precision, recall, and F1-score. The highest accuracy value of 99.17% was obtained when the light gradient boosting machine classifier was used together with the SMOTE data augmentation technique. It was observed that the measurements obtained in the study were higher than the results obtained in similar studies conducted in the literature in recent years. [ABSTRACT FROM AUTHOR]

Support-vector machines (SVMs) have been successfully employed to diagnose neurological disorders like epilepsy and sleep disorders via electroencephalogram (EEG) signal classification. However, EEG signals are susceptible to noise and outliers while recording. Thus, classification of EEG signals require efficient methods that is robust to noise and outliers. SVMs suffer in the presence of noise and outliers as it treats each sample to be of equal importance. Fuzzy membership function has been successful in dealing with noise and outliers. Fuzzy support vector machine uses fuzzy membership to give the appropriate weight to each sample for reducing the effect of noise and outliers. To reduce the effect of noise and outliers, intuitionistic fuzzy twin support vector machines (IFTWSVM) uses both membership and non-membership weights to subsidise the effect of outliers. Moreover, IFTWSVM solves smaller size quadratic programming problems which makes it comparatively efficient than fuzzy support vector machines. However, IFTWSVM suffers as (i) it involves the computation of matrix inverses which makes it intractable as the size of the data increases. (ii) it solves different problems for linear and nonlinear cases, and nonlinear case with linear kernel does not reduce to the linear case. Thus, it can only solve the approximate formulation not the exact formulation. To overcome these issues, we propose improved intuitionistic fuzzy twin support vector machine (IIFTWSVM). The proposed IIFTWSVM introduces different Lagrangian functions to avoid the computation of matrix inverses. Also, in nonlinear case, kernel trick is applied directly and hence, the exact formulation is solved. We employed the proposed IIFTWSVM for the classification of EEG signals. The experimental results and statistical analysis show that the proposed IIFTWSVM model is better compared to the given baseline models. Also, we evaluate the performance of the models on KEEL datasets to check the robustness of the proposed IIFTWSVM model. [ABSTRACT FROM AUTHOR]

Twin support vector regression (TSVR) has been widely applied in regression problems. TSVR seeks a pair of ε -insensitive proximal planes by solving two support vector machine type problems. TSVR assumes that the matrices appearing in the dual formulation are positive definite. However, in real-world scenarios, such an assumption may not be fulfilled and, hence, leads to suboptimal performance. ε -Twin support vector regression (ε -TSVR) improved the TSVR by introducing the regularisation term to avoid the singularity issues. Most of the twin support vector machine models involve the computation of matrix inverses. Also, TSVR implements the empirical risk minimization principle. In this paper, we propose an improved twin support vector regression (ITSVR) for brain age estimation by introducing different Lagrangian functions for the primal problems of the TSVR. The proposed ITSVR implements the structural risk minimization principle and avoids the computation of the matrix inverses. To solve the optimization problem more efficiently, we used successive overrelaxation (SOR) technique. We evaluated the proposed ITSVR on cognitively healthy subjects, mild cognitive impairment subjects and Alzheimer's disease subjects. The experimental results demonstrate that the proposed ITSVR has superior performance compared to the baseline models for brain age estimation. [ABSTRACT FROM AUTHOR]

With the aging of the population, China has a large population base. The number of people suffering from mild cognitive impairment has gradually increased and gradually turned to senile dementia. The probability of mild cognitive impairment in China is about 5–7%, and about 15 million people are ill. To better distinguish mild cognitive impairment from Alzheimer's, this paper adopted the machine learning (ML) method to model. The evaluation table was established by considering the reaction time, education, background, memory, and other aspects of the patient. Machine learning has been applied in the fields of the Internet, finance, medicine, automation, biological science, etc. Through the machine learning support vector machine (SVM) and linear regression, the classification accuracy was compared. The results showed that the classification accuracy of SVM was 87.92% under the ML algorithm. It also showed that ML was more conducive to classification and recognition, which has played an important role in the identification of mild cognitive impairment in the current aging population. [ABSTRACT FROM AUTHOR]

Protein and carbohydrate content in the diet of the predator Hippodamia variegata Goeze (variegated ladybug) directly influences reproduction fitness by affecting foraging efficiency. The effect of three varied qualities of Aphis gossypii Glover (cotton aphid) prey on daily H. variegata feeding and daily egg production (DEP) were estimated using the support vector machine (SVM). The SVM can predict predator reproductive behavior as a function of the relationship between foraging and prey nutritional composition. We used the total number and weight of aphids consumed, volume of protein, lipid, carbohydrate, and glycogen, and total energy received by the ladybug after feeding on one prey item as input variables. Aphid quality varied as nitrogen (N) fertilization levels were 110, 160, and 210 ppm on the Cucumis sativus L. (cucumber) host plants. The model estimated female beetles consumed more aphids and nutrients on C. sativus plants with N levels of 160 ppm compared to lower (110 ppm) N levels and had higher reproductive transformation efficiencies. Transformation rates of aphid feeding to egg production in females exposed to the 160 ppm treatment were 65% greater, had lower nutrient and energy requirements, and achieved a 29.4% higher DEP than those exposed to high (210 ppm) N levels. The SVM predicted nutrient compositions of A. gossypii exposed to 160 ppm N were balanced such that H. variegata exhibited greater reproduction efficiency than the other N treatments for first 30 d from the start of reproduction. [ABSTRACT FROM AUTHOR]

Breast cancer is a significant global health concern, with millions of cases and deaths each year. Accurate diagnosis is critical for timely treatment and medication. Machine learning techniques have shown promising results in detecting breast cancer. Previous studies have primarily used single-modality data for breast cancer diagnosis. Hence, this work aims to mobilize the benefits of multimodal data over unimodality samples. This study proposes a custom deep learning-based model pipeline that works over this multimodal data. This work has been separated into three phases. Phase 1 and Phase 2 under the unimodal category examine gene expression data and histopathological images separately. The Cancer Genome Atlas makes these datasets available. In Phase 3, the proposed pipeline operates on both data types' samples for each patient in the multimodal category. This study investigates how data pre-processing (cleaning, transformation, reduction) and cascaded filtering affect model performance. Precision, recall, f1-score, and accuracy assessed the models, whereas L2 regularization, exponentially weighted moving average, and transfer learning minimized over-fitting. A custom deep neural network and support vector machine obtained 86% accuracy in Phase 1, whereas the VGG16 model reached 80.21% accuracy in Phase 2. In Phase 3, the curated multimodal dataset was applied to a custom deep learning pipeline (VGG16 backbone with hyper-tuned machine learning models as head classifiers) to achieve 94% accuracy, demonstrating the importance of multimodal data over unimodal in breast cancer subtype classification. These findings highlight the importance of multimodal data for breast cancer diagnosis and subtype prediction. [ABSTRACT FROM AUTHOR]

Mental task classification (MTC), based on the electroencephalography (EEG) signals is a demanding brain–computer interface (BCI). It is independent of all types of muscular activity. MTC-based BCI systems are capable to identify cognitive activity of human. The success of BCI system depends upon the efficient feature representation from raw EEG signals for classification of mental activities. This paper mainly presents on a novel feature representation (formation of most informative features) of the EEG signal for the both, binary as well as multi MTC, using a combination of some statistical, uncertainty and memory- based coefficient. In this work, the feature formation is carried out in the two stages. In the first stage, the signal is split into different oscillatory functions with the help of three well-known empirical mode decomposition (EMD) algorithms, and a new set of eight parameters (features) are calculated from the oscillatory function in the second stage of feature vector construction. Support vector machine (SVM) is used to classify the feature vectors obtained corresponding to the different mental tasks. This study consists the problem formulation of two variants of MTC; two-class and multi-class MTC. The suggested scheme outperforms the existing work for the both types of mental tasks classification. [ABSTRACT FROM AUTHOR]

With more than 50 million people worldwide at risk of heart disease, early diagnosis of cardiovascular disease is essential. The classification of electrocardiogram (ECG) recordings can diagnose Atrial Fibrillation (AF) and other types of arrhythmia. Short ECG segments recorded from wearable devices in an IoT-based system can further provide continuous abnormality detection. The performance of ECG classification is usually high in normal segments, but much lower in the target classes, i.e., AF and other arrhythmias, which could result from class imbalance and limited feature representation. Deep learning methods have been employed as feature extractors in previous studies. Among them, convolutional neural networks (CNN) can generate rich features in different scales. But CNN may omit precise temporal information such as the duration between R-waves in two QRS waves (RR interval) irregularity, which is insensitive to noise segments. Thus, aiming at improving the classification performance of AF and other classes in short ECG segments, we propose a hybrid feature fusion method integrating the above-mentioned features. The fused features are trained and tested in a support vector machine classifier. The F1-score results show that our method outperforms not only the same CNN method without feature fusion in all the four classes, which average F1-score reached 84.3% and classification time per single sample of 0.005 s, but also several state-of-the-art methods, especially in the target classes, which validates the effectiveness of the proposed method. We then further discuss the impact of length on the performance of the proposed method, providing insights into future applications. [ABSTRACT FROM AUTHOR]

In the field of nanoscience, the scanning electron microscope (SEM) is widely employed to visualize the surface topography and composition of materials. In this study, we present a novel SEM image classification model called NFSDense201, which incorporates several key components. Firstly, we propose a unique nested patch division approach that divides each input image into four patches of varying dimensions. Secondly, we utilize DenseNet201, a deep neural network pretrained on ImageNet1k, to extract 2920 deep features from the last fully connected and global average pooling layers. Thirdly, we introduce an iterative neighborhood component analysis function to select the most discriminative features from the merged feature vector, which is formed by concatenating the four feature vectors extracted per input image. This process results in a final feature vector of optimal length 698. Lastly, we employ a standard shallow support vector machine classifier to perform the actual classification. To evaluate the performance of NFSDense201, we conducted experiments using a large public SEM image dataset. The dataset consists of 972, 162, 326, 4590, 3820, 3925, 4755, 181, 917, and 1624.jpeg images belonging to the following microstructural categories: "biological," "fibers," "film-coated surfaces," "MEMS devices and electrodes," "nanowires," "particles," "pattern surfaces," "porous sponge," "powder," and "tips," respectively. For both four-class and ten-class classification tasks, we evaluated NFSDense201 using subsets of the dataset containing 5080 and 21,272 images, respectively. The results demonstrate the superior performance of NFSDense201, achieving a four-class classification accuracy rate of 99.53% and a ten-class classification accuracy rate of 97.09%. These accuracy rates compare favorably against previously published SEM image classification models. Additionally, we report the performance of NFSDense201 for each class in the dataset. [ABSTRACT FROM AUTHOR]

Colorectal cancer (CRC) is one of the most lethal kinds of cancer, so early detection is critical. Three datasets, namely CNN transfer learning with discrete wavelet transform (DWT), discrete cosine transform (DCT), and support vector machines (SVMs), were used to find CRC. In these instances, a quick and precise visual diagnosis of polyps is needed in the current scenario. The proposed process involves four distinct phases. First and foremost, convolutional neural networks (CNNs) are developed to test the efficacy of the model. Further, a transfer learning approach was incorporated using SVM and LSTM. Using the K-means technique, a visual explanation is finally presented. This system works with the balanced Hyper Kvasir and mixed datasets, which are made up of CVC Clinic DB, Kvasir2, and Hyper Kvasir. The system is called "ColoRectalCADx". The convolutional neural network (CNN) models are ResNet-50V2, DenseNet-201, VGG-16, and RDV-22. The system achieved the highest accuracy with CNN DesnseNet-201 in Hyper Kvasir (98.92% training, 98.91% testing, 93.62% SVM training, and 95.87% SVM tests). CNN DenseNet-201 also achieved the highest accuracy with the mixed dataset (98.91% training, 96.13% testing, 95.41% SVM training, and 94.86% SVM testing). The process involved three phases, namely individual CNN, combination of CNN with SVM, and combination of CNN, LSTM, and SVM. After three phases of the system, across both datasets, the CNN + SVM + LSTM combination was proven to be the most effective. Finally, the unsupervised K-means learning algorithm extracts the location of any cancerous polyps and upon classification using SVM classifier resulted with an accuracy of 80%. The K-means algorithm, which uses segmented images as input, accurately predicts the sites of tumours in colorectal cancer patients. [ABSTRACT FROM AUTHOR]

The selection of proper classifiers for a given data set is full of challenges. The critical problem of classifier selection is how to extract feature from data sets. This paper proposes a new method for feature extraction of a data set. Our method not only preserves the geometrical structure of a data set, but also characterizes the decision boundary of classification problems. Specifically speaking, the extracted feature can recover a data set that has the same Euclidean geometrical structure as the original data set. We present an efficient algorithm to compute the similarity between data set features. We theoretically analyze how the similarity between our features affects the performance of the support vector machine, a well-known classifier. The empirical results show that our method is effective in finding suitable classifiers. [ABSTRACT FROM AUTHOR]

Coconut water is the clear liquid found inside coconuts, famous for rehydrating after exercise or while suffering from a minor sickness. The essential issue tackled in this paper is how to estimate the appropriate stage of maturity of coconut water, which is a time-consuming task in the beverage industry since, as the coconut age increases, the coconut water flavor varies. Accordingly, to handle this issue, an adaptive model based on Fuzzy Neural Network and Sperm Whale Optimization, dubbed FNN–SWO, is developed to assess coconut water maturity. The Sperm Whale Optimization (SWO) algorithm is a meta-heuristic optimization algorithm. It is embedded in this model along with neural networks and fuzzy techniques (FNN system), which can be employed as an essential building block in the beverage industry. The proposed FNN–SWO model is trained and tested utilizing fuzzy rules with an adaptive network. In contrast, the SWO algorithm is adopted to determine the optimal weights for the fuzzy rules. Three subsets of data divided according to three levels of coconut water maturity-tender, mature, and very mature, are used to validate the combined FNN–SWO model. Depending on these three subsets of data, a comparison of the proposed FNN–SWO model has been conducted against a set of the most common conventional techniques. These techniques include Support Vector Machine, Naïve Bayes, FNN, Artificial Neural Network, as well as their embedding with other meta-heuristic optimization algorithms. For various key performance indicators, such as recall, F1-score, specificity, and accuracy, the proposed FNN–SWO model provides the best prediction outcomes compared to the current time-consuming techniques. The dominance of the proposed FNN–SWO model is evident from the final findings compared to its time-consuming peers for estimating coconut water maturity on time. As a result, the proposed FNN–SWO model is an effective heuristic for locating optimal solutions to classification problems. It can thereby be reassuringly applicable to other similar prediction problems. Additionally, it would benefit the scientific community interested in evaluating coconut water. [ABSTRACT FROM AUTHOR]

Accurate short-term electricity load forecasts are critical for the secure and economic operation of power systems. This paper presents a computationally efficient and powerful three-stage model to accurately forecast short-term electricity load. Variational mode decomposition (VMD) was used in the first stage to extract features from the historical load signal. The stacked kernel extreme learning machine (KELM)-based auto-encoders were utilized in the unsupervised feature learning-based second stage. In the third stage, the high-order learned features were used as the inputs for the KELM-based regression model. The proposed deep KELM architecture combines stacked KELM-based auto-encoders and a KELM-based regression model to forecast short-term electricity load effectively. In order to examine the performance improvement of the proposed forecasting model, several performance comparison tests were realized using publicly available electricity load and day-ahead forecast data from the Turkish transmission system operator (TSO). The proposed model results were compared with state-of-the-art deep extreme learning machine (ELM) architectures as well as the benchmark forecasting models based on original ELM, KELM, artificial neural network (ANN), support vector machine (SVM), and regression tree (RT). The comparison results indicated that the proposed model outperformed state-of-the-art architectures and was significantly more successful than the TSO model. [ABSTRACT FROM AUTHOR]

Recently, various techniques have been implemented to investigate the cavity beneath the pavement using ground-penetrating radar (GPR) data. These techniques usually focus on the determination of the height of the cavity and the depth at which it is located, which play a vital role in the agencies' decision-making on the appropriate rehabilitation method for the pavement. Image processing, a relatively advanced technique that utilizes the image generated from the GPR data, has been used for this purpose. However, it is still time-consuming and yields low accuracy. As a solution, this study proposes a machine learning (ML)-based method to estimate cavity properties, such as depth and height, from GPR images with high precision in a significantly reduced time. In this study, a deep convolutional generative adversarial networks-based hyperbolic fitting algorithm, called GAN–cavity, was developed based on the Pix2Pix model to identify multi-shape signature segmentation from the GPR data. The GAN–cavity was trained and tested using 1920 and 480 images, respectively, and the model performed well with an average F1–score of 83.1 per cent. Consequently, an extreme gradient boosting (XGBoost) algorithm was used to predict the cavity depth and height based on the GAN–cavity results. Moreover, the developed model (XGB–cavity classifier) can classify the cavity's height into three classes: small, medium, and large. The XGBoost model XGB–cavity regressor used 341 and 87 cavities for training and testing, respectively, yielding regression scores (R2) and root mean squared error (RMSE) of 0.88 and 3.28 cm in cavity depth prediction, respectively. By applying the support vector machine synthetic minority oversampling technique (SVM–SMOTE) algorithm to overcome the data imbalance, the XGB–cavity classifier model obtained a high accuracy of 0.89 with precision–avg, recall–avg, and F1–score–avg of 0.83, 0.83, and 0.83, respectively. Based on the results, the proposed method (GAN–XGB–cavity) of combining the GAN–cavity and XGB–cavity can successfully identify and fit multi-shape segmentation for depth estimation and height classification of cavities. [ABSTRACT FROM AUTHOR]

In a smart grid, the main goals are to provide grid stability, improve power system performance and security, and reduce operations, system maintenance, and planning costs. The prediction stability of smart grid (SG) systems is essential in terms of power loss minimization and the importance of adequate energy policies. SG systems must accurately predict the energy demand and ensure the right amount of energy is available at the right time. If the prediction is inaccurate, it can lead to costly energy production or usage errors and create considerable inefficiencies in the power grid. Due to this, this manuscript offers five different cascade methods to detect the stability of SG systems. Detecting the stability of SG systems enables the grid to respond quickly to changes in demand and supply, improves system reliability, reduces power outages, and increases the overall efficiency of the grid. The present work proposed five different cascade methods with pre-processing, training and testing division, and the classification stages of the classification procedure for estimating SG stability. In the first pre-processing stage, the SG dataset is pre-proceeded with the feature selection (Relief, Fast Correlation-Based Filter (FCBF), and supervised attribute filter). The resampling (the bootstrapping), the Fuzzy C-Means Clustering-Based Feature Weighting (FCMFW), the resampling then feature selection (supervised attribute filter), and the feature selection (supervised attribute filter), then FCMFW. In the second stage, the training and testing division stage, the SG dataset was separated into three test and training data methods before the classification algorithm: The 5 Fold Cross Validation (FVC), 10 FVC, and hold-out (50–50%). In the third stage, the classification stage, five different classification algorithms, including Naive Kernel Bayes, Linear Support Vector Machine (SVM), Weighted K-Nearest Neighbors, Begged Trees, and Narrow Neural Network classifying algorithms, are used to classify the SG dataset. The simulation results of this study demonstrated that the suggested cascade ML system had achieved significant accuracy in predicting SG stability. The best cascade method is the feature selection (supervised attribute filter) + FCMFW + 10 FCV and then performing the bagged trees algorithm; thus, the new approach affords an accuracy of 99.9%. Furthermore, due to the rapid growth of ML techniques, sensors, and smart meters technologies, with Machine to Machine communication via the internet of things (IoT), the real-time identification process is made practical with higher accuracy. For this reason, our future research will focus on an IoT-based SG system, an E-stability determination system. Thanks to the proposed cascade method, the SG dataset can be classified easily, quickly, and reliably. E-stability determination systems can help to fast detect, predict, and respond, which is an important application of IoT on the grid systems. [ABSTRACT FROM AUTHOR]

Weirs are hydraulic structures mostly used to measure the flow discharge and control the flow level in artificial or natural open channels. The ratio of the actual discharge to the theoretical discharge (discharge coefficient—Cd) must be known, in order to calculate the discharge of the channel having the weir. In this study, 91 experimental measurements are taken on seven trapezoidal broad-crested weirs with different upstream and downstream slopes. Experimentally measured flow properties are used to validate numerical models based on the computational fluid dynamics (CFD) methods. Two new weir geometries, not experimentally measured, are added in the numerical modeling, and 270 Cd values are calculated for nine weir geometries using numerical modeling. Theoretical Cd values are estimated using the artificial neural network (ANN), support vector machine (SVM), and M5Tree methods. In the models, the Froude number in the upstream region and dimensionless parameters of the flow are used as inputs. The performance of these methods has been examined to estimate the Cd values for eight cases. The performances of the methods are evaluated by the coefficient of determination (R2), root-mean-square error, mean absolute percentage error, and Nash–Sutcliffe model efficiency coefficient. The study results show that the Froude number significantly increases the performance of the models in estimating Cd values, and the ANN method is more successful in determining Cd than other methods. [ABSTRACT FROM AUTHOR]

With the increased use of social network sites, such as Twitter, attackers exploit these platforms to spread counterfeit content. Such content can be fake advertisements or illegal content. Classifying such content is a challenging task, especially in Arabic. The Arabic language has a complex structure and makes classification tasks more difficult. This paper presents an approach to classifying Arabic tweets using classical machine learning (non-deep machine learning) and deep learning techniques. Tweets corpus were collected through Twitter API and labelled manually to get a reliable dataset. For an efficient classifier, feature extraction is applied to the corpus dataset. Then, two learning techniques are used for each feature extraction technique on the created dataset using N-gram models (uni-gram, bi-gram, and char-gram). The applied classical machine learning algorithms are support vector machines, neural networks, logistics regression, and naïve Bayes. Global vector (GloVe) and fastText learning models are utilised for the deep learning approaches. The Precision, Recall, and F1-score are the suggested performance measures calculated in this paper. Afterwards, the dataset is increased using the synthetic minority oversampling technique class to create a balanced dataset. After applying the classical machine learning models, the experimental results show that the neural network algorithm outperforms the other algorithms. Moreover, the GloVe outperforms the fastText model for the deep learning approach. [ABSTRACT FROM AUTHOR]

The operation of large utility companies such as Consolidated Edison Company of New York, Inc. (Con Edison) typically rely on large quantities of regulation documents from external institutions which inform the company of upcoming or ongoing policy changes or new requirements the company might need to comply with if deemed applicable. As a concrete example, if a recent regulatory publication mentions that the timeframe for the Company to respond to a reported system emergency in its service territory changes from within X time to within Y time—then the affected operating groups will be notified, and internal Company operating procedures may need to be reviewed and updated accordingly to comply with the new regulatory requirement. Each such regulation document needs to be reviewed manually by an expert to determine if the document is relevant to the company and, if so, which department it is relevant to. In order to help enterprises improve the efficiency of their operation, we propose an automatic document classification pipeline that determines whether a document is important for the company or not, and if deemed important it forwards those documents to the departments within the company for further review. Binary classification task of determining the importance of a document is done via ensembling the Naive Bayes (NB), support vector machine (SVM), random forest (RF), and artificial neural network (ANN) together for the final prediction, whereas the multi-label classification problem of identifying the relevant departments for a document is executed by the transformer-based DocBERT model. We apply our pipeline to a large corpus of tens of thousands of text data provided by Con Edison and achieve an accuracy score over 80 % . Compared with existing solutions for document classification which rely on a single classifier, our paper i) ensemble multiple classifiers for better accuracy results and escaping from the problem of overfitting, ii) utilize pretrained transformer-based DocBERT model to achieve ideal performance for multi-label classification task and iii) introduce a bi-level structure to improve the performance of the whole pipeline where the binary classification module works as a rough filter before finally distributing the text to corresponding departments through the multi-label classification module. [ABSTRACT FROM AUTHOR]

For semi-supervised learning, a few labeled data and a large number of unlabeled data are used to construct a reasonable classifier. In recent years, many semi-supervised learning methods have been proposed and achieved good performance, especially for the graph-based approaches that can exploit the geometric information embedded in the data. Motivated by the success of generalized elastic net Lp-norm nonparallel support vector machine (GLpNPSVM) and the graph-based regularization term, in this paper, a novel Laplacian generalized elastic net Lp-norm nonparallel support vector machine for semi-supervised learning (Lap-GLpNPSVM) is proposed. A Lp-norm graph regularization term is introduced to improve the performance by the adjustability of the value of p. Experimental results on two synthetic datasets, fifteen UCI datasets, and Handwritten Numeral datasets demonstrate that proposed methods outperform other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Assessing the quality of road surfaces for maintenance has typically been an expensive exercise for local government agencies. This paper evaluates the feasibility of using accelerometer and speed data collected from on-board diagnostic (OBD-II) devices and labelled using the PASER system of road classification to classify road quality and identify poor stretches. An ordinal logistic model and a support vector machine (SVM) classifier were first trained on features created from raw data. The SVM and an artificial neural network (ANN) were then trained separately on the raw data parameters followed by a convolutional neural network (CNN) model. All models have comparable results in terms of overall classification accuracy on validation datasets (around 67%) but the ANN and CNN models were superior in their ability to correctly identify 'Poor' and 'Good' stretches. The CNN model had the best performance among all models evaluated. The approach presented offers a low-cost method to map poor roads and identify stretches that require more attention as an alternative to more expensive traditional methods. [ABSTRACT FROM AUTHOR]

Motor imagery (MI) brain–computer interface (BCI) systems have broad application prospects in rehabilitation and other fields. However, to achieve accurate and practical MI-BCI applications, there are still several critical issues, such as channel selection, electroencephalogram (EEG) feature extraction and EEG classification, needed to be better resolved. In this paper, these issues are studied for lower limb MI which is more difficult and less studied than upper limb MI. First, a novel iterative EEG source localization method is proposed for channel selection. Channels FC1, FC2, C1, C2 and Cz, instead of the commonly used traditional channel set (TCS) C3, C4 and Cz, are selected as the optimal channel set (OCS). Then, a multi-domain feature (MDF) extraction algorithm is presented to fuse single-domain features into multi-domain features. Finally, a particle swarm optimization based support vector machine (SVM) method is utilized to classify the EEG data collected by the lower limb MI experiment designed by us. The results show that the classification accuracy is 88.43%, 3.35–5.41% higher than those of using traditional SVM to classify single-domain features on the TCS, which proves that the combination of OCS and MDF can not only reduce the amount of data processing, but also retain more feature information to improve the accuracy of EEG classification. [ABSTRACT FROM AUTHOR]

The coronavirus pandemic has been globally impacting the health and prosperity of people. A persistent increase in the number of positive cases has boost the stress among governments across the globe. There is a need of approach which gives more accurate predictions of outbreak. This paper presents a novel approach called diffusion prediction model for prediction of number of coronavirus cases in four countries: India, France, China and Nepal. Diffusion prediction model works on the diffusion process of the human contact. Model considers two forms of spread: when the spread takes time after infecting one person and when the spread is immediate after infecting one person. It makes the proposed model different over other state-of-the art models. It is giving more accurate results than other state-of-the art models. The proposed diffusion prediction model forecasts the number of new cases expected to occur in next 4 weeks. The model has predicted the number of confirmed cases, recovered cases, deaths and active cases. The model can facilitate government to be well prepared for any abrupt rise in this pandemic. The performance is evaluated in terms of accuracy and error rate and compared with the prediction results of support vector machine, logistic regression model and convolution neural network. The results prove the efficiency of the proposed model. [ABSTRACT FROM AUTHOR]

With each passing day, the number of smart vehicles is increasing manifold, hence, automatic/automated parking lot detection is gaining a lot of importance among Smart City applications. A robust approach is desired to identify parking spaces effectively and efficiently. This work presents a deep learning classifier based on convolutional neural network (CNN) and extreme learning machine (ELM), i.e., CNN-ELM to classify the parking space as vacant or occupied. CNN is well-known for efficient image classification, but its training time is highly influenced by backpropagation of errors in the fully connected layer. Thus, ELM is plugged into CNN to replace the fully connected layer and perform classification whereas, feature extraction is performed using CNN. The performance of CNN-ELM is validated on the publicly available PKLot dataset, which contains approximately 700,000 images categorized into sunny, overcast, and rainy weather conditions. The experimental results indicate that CNN-ELM approach outperforms other hybrid CNN models using different classifiers such as support vector machine, Xgboost, and Extra Trees in terms of sensitivity, specificity, and accuracy. The comparison of results with other state-of-the-art approaches based on accuracy and Area under the curve (AUC) score further justifies the effectiveness of the proposed approach in real-time parking space detection. [ABSTRACT FROM AUTHOR]

As a semi-supervised learning method, Laplacian support vector machine (LapSVM) is popular. Unfortunately, the model generated by LapSVM has a poor sparsity. A sparse decision model has always been fascinating because it could implement data reduction and improve performance. To generate a sparse model of LapSVM, we propose an ℓ 1 -norm Laplacian support vector machine ( ℓ 1 -norm LapSVM), which replaces the ℓ 2 -norm with the ℓ 1 -norm in LapSVM. The ℓ 1 -norm LapSVM has two techniques that can induce sparsity: the ℓ 1 -norm regularization and the hinge loss function. We discuss two situations for the ℓ 1 -norm LapSVM, linear and nonlinear ones. In the linear ℓ 1 -norm LapSVM, the sparse decision model implies that features with nonzero coefficients are contributive. In other words, the linear ℓ 1 -norm LapSVM can perform feature selection to achieve the goal of data reduction. Moreover, the nonlinear (kernel) ℓ 1 -norm LapSVM can also implement data reduction in terms of sample selection. In addition, the optimization problem of the ℓ 1 -norm LapSVM is a convex quadratic programming one. That is, the ℓ 1 -norm LapSVM has a unique and global solution. Experimental results on semi-supervised classification tasks have shown a comparable performance of our ℓ 1 -norm LapSVM. [ABSTRACT FROM AUTHOR]

Epilepsy is one of the most common neurological diseases, affecting approximately 50 million people. This illness can be diagnosed by electroencephalogram (EEG), whose analysis depends on human interpretation, which may lead to divergent results and be imprecise and error-prone. Moreover, one estimates more than 80% of the epilepsy examinations return no anomalies at all, wasting the effort of analysis. In this context, machine learning (ML) methods are used to aid in epilepsy detection. An essential task for classifier building is feature extraction, for which there are many measures that can be computed. However, part of these features can be irrelevant or hinder the performance of ML methods. This paper proposes an automatic way to classify EEG segments by using a reduced set of features. The proposed method combines multispectral analysis, feature selection, and classifier building approaches. As our main contribution, a methodology to minimize the number of features for classifier building is proposed. A total of 285 measures are extracted. Afterward, two attribute selection approaches are used: Pearson's correlation coefficient filter and wrapper based on the genetic algorithm. Then, five well-known ML methods (k-nearest-neighbor, support vector machine, naive Bayes, artificial neural network, and random forest) were used to build 281 different classifiers. As a result, the proposed classifiers reached an accuracy between 87.2 and 90.99% and considerably reduced the number of features from 285 to 30, keeping competitive scores. Additionally, statistical hypothesis tests prove that our proposed approach is as efficient as using the complete feature dataset for classifier building. [ABSTRACT FROM AUTHOR]

Deep convolutional neural networks (DCNNs) are one of the most advanced techniques for classifying images in a range of applications. One of the most prevalent cancers that cause death in women is breast cancer. For survival rates to increase, early detection and treatment of breast cancer is essential. Deep learning (DL) can help radiologists diagnose and classify breast cancer lesions. This paper proposes a computer-aided system based on DL techniques for automatically classify breast cancer tumors in histopathological images. There are nine DCNN architectures used in this work. Four schemes are performed in the proposed framework to find the best approach. The first scheme consists of pre-trained DCNNs based on the transfer learning concept. The second scheme performs feature extraction of the DCNN architectures and uses a support vector machine (SVM) classifier for evaluation. The third one performs feature integration to show how the integrated deep features may enhance the SVM classifiers' accuracy. Finally, in the fourth scheme, the Chi-square (χ2) feature selection method is applied to reduce the large feature size in the feature integration step. The results of the proposed system present a promising performance for breast cancer classification with an accuracy of 99.24%. The system performance shows that the proposed tool is suitable to assist radiologists in diagnosing breast cancer tumors. [ABSTRACT FROM AUTHOR]

One of the most important problems for farmers who produce large amounts of apples is the classification of the apples according to their types in a short time without handling them. Support vector machines (SVM) and deep residual networks (ResNet-50) are machine learning methods that are able to solve general classification situations. In this study, the classification of apple varieties according to their genus is made using machine learning algorithms. A database is created by capturing 120 images from six different apple species. Bag of visual words (BoVW) treat image features as words representing a sparse vector of occurrences over the vocabulary. BoVW features are classified using SVM. On the other hand, ResNet-50 is a convolutional neural network that is 50 layers deep with embedded feature extraction layers. The pre-trained ResNet-50 architecture is retrained for apple classification using transfer learning. In the experiments, our dataset is divided into three cases: Case 1: 40% train, 60% test; Case 2: 60% train, 40% test; and Case 3: 80% train, 20% test. As a result, the linear, Gaussian, and polynomial kernel functions used in the BoVW + SVM algorithm achieved 88%, 92%, and 96% accuracy in Case 3, respectively. In the ResNet-50 classification, the root-mean-square propagation (rmsprop), adaptive moment estimation (adam), and stochastic gradient descent with momentum (sgdm) training algorithms achieved 86%, 89%, and 90% accuracy, respectively, in the set of Case 3. [ABSTRACT FROM AUTHOR]

A computerized coconut tree detection system can help dendrologists and laypersons in identifying coconut trees based on three morphological parameters including height, inclination, and orientation. These three parameters help to determine the health and the nature of growth of coconut trees which influences the design and use of robots for harvesting coconuts. Deep learning is a powerful tool used for feature extraction as it is better in extracting deeper details (features) in an image. In this research work, a new Modified Inception Net based Hyper Tuning Support Vector Machine classification method named MIN-SVM is proposed for coconut tree classification based on three morphological parameters including height, inclination and orientation. The features from the pre-processed coconut tree images were extracted using four distinct Convolutional Neural Network models including Visual Geometry Group, Inception Net, ResNet, and MIN-SVM. These extracted features were then classified using a Machine Learning model named Support Vector Machine (SVM). The MIN-SVM have achieved a remarkable accuracy of 95.35 percent as contrasted to Visual Geometry Group (91.90%), Inception Net (81.66%), and ResNet (71.95%). The features extracted from Modified Inception Net fitted good with SVM classifier. Experimental results show that MIN-SVM can be powerful computerized automated system to identify coconut trees based on height, inclination, and orientation. [ABSTRACT FROM AUTHOR]

In this manuscript, imbalanced and small sample space (IB-SSS) dataset problems for pedestrian gender classification using fusion of selected deep and traditional features (PGC-FSDTF) are considered. In this regard, data preparation is first done through data augmentation and preprocessing steps to handle imbalanced classification problem and environmental effects, respectively. The proposed approach follows different types of feature extraction schemes, for instance, pyramid histogram of oriented gradients, hue saturation value histogram, deep visual features of DenseNet201 and InceptionResNetV2-based convolutional neural network architectures. The parallel fusion method computes the maximum and average values-based features from the learned features of both deep networks. Features are selected through features selection methods such as entropy and principal component analysis (PCA). The subsets of features are serially fused and provided to multiple classifiers to perform gender classification on IB-SSS datasets. Resultantly, the proposed PGC-FSDTF method shows better results in terms of different accuracies (overall, mean, and balanced), and area under curve on selected datasets. Further, improved results are achieved on applied datasets using PCA-based selected features and medium Gaussian support vector machine (M-SVM) classifier. These results on different datasets confirm that the selected feature combination provides a way to handle IB-SSS issues for PGC effectively. [ABSTRACT FROM AUTHOR]

Academic performance, a globally understood metric, is utilized worldwide across disparate teaching and learning environments and is regarded as a quantifiable indicator of learning gain. The ability to reliably estimate student's academic performance is important and can assist academic staff to improve the provision of support. However, it is recognized that academic performance estimation is non-trivial and affected by multiple factors, including a student's engagement with learning activities and their social, geographic, and demographic characteristics. This paper investigates the opportunity to develop reliable models for predicting student performance using Artificial Intelligence. Specifically, we propose two-step academic performance prediction using feature weighted support vector machine and artificial neural network (ANN) learning. A feature weighted SVM, where the importance of different features to the outcome is calculated using information gain ratios, is employed to perform coarse-grained binary classification (pass, P 1 , or fail, P 0 ). Subsequently, detailed score levels are divided from D to A+, and ANN learning is employed for fine-grained, multi-class training of the P 1 and P 0 classes separately. The experiments and our subsequent ablation study, which are conducted on the student datasets from two Portuguese secondary schools, have proved the effectiveness of this hybridized method. [ABSTRACT FROM AUTHOR]

Discovering a hearing disorder at an earlier intervention is critical for reducing the effects of hearing loss and the approaches to increase the remaining hearing ability can be implemented to achieve the successful development of human communication. Recently, the explosive dataset features have increased the complexity for audiologists to decide the proper treatment for the patient. In most cases, data with irrelevant features and improper classifier parameters causes a crucial influence on the audiometry system in terms of accuracy. This is due to the dependent processes of these two, where the classification accuracy performance could be worsened if both processes are conducted independently. Although the filter algorithm is capable of eliminating irrelevant features, it still lacks the ability to consider feature reliance and results in a poor selection of significant features. Improper kernel parameter settings may also contribute to poor accuracy performance. In this paper, an ensemble filters feature selection based on Information Gain (IG), Gain Ratio (GR), Chi-squared (CS), and Relief-F (RF) with harmonize optimization of Particle Swarm Optimization (PSO) and Support Vector Machine (SVM) is presented to mitigate these problems. Ensemble filters are utilized so that the initial top dominant features relevant for classification can be considered. Then, PSO and SVM are optimized simultaneously to achieve the optimal solution. The results on a standard Audiology dataset show that the proposed method produces 96.50% accuracy with optimal solution compared to classical SVM, which signifies the proposed method is effective in handling high dimensional data for hearing disorder prediction. [ABSTRACT FROM AUTHOR]

Accurate streamflow estimation is crucial for proper water management for irrigation, hydropower, drinking and industrial purposes. The main aim of this study to adopt new data preprocessing techniques (e.g., EMD, EEMD and EWT) to capture the data noise and to enhance the prediction accuracy of machine learning methods for streamflow estimation which is a challenging task in high-altitude basins due to the influence of many external climatic and geographical parameters. The prediction accuracy of support vector regression (SVR), twin support vector machine (T), extreme learning machine (ELM), asymmetric Huber loss function-based ELM (AHELM) and ε-insensitive Huber loss function-based ELM (ε-AHELM) methods are investigated in monthly streamflow prediction. Among the standalone methods, the ε-AHELM performs superior to the SVR, TSVR, ELM, and AHELM in streamflow prediction; improvements in root mean square error are 6.9%, 4.9%, 6% and 4.2%, respectively. The study outcomes reveal that the preprocessing methods considerably improve the prediction accuracy of the implemented standalone models. Among the data preprocessing techniques, it is found that the EWT outperforms the EMD and EEMD techniques by reducing the prediction errors in the best ε-AHELM, EMD-ε-AHELM and EEMD-ε-AHELM models by 68–61.3%, 64.7–63.4% and 59.4–58.6%, respectively. The overall results of the study recommend the use of EWT-ε-AHELM in streamflow estimation. [ABSTRACT FROM AUTHOR]

In this study, a new deep learning-based approach has been developed that detects and classifies surface defects that occur in the steel production process. The proposed methodology was created in four steps. In the first step, a deep learning model is designed that trains the residual and attention structures in parallel, thus increasing the classification performance. In the second step, deep features were extracted from the Parallel Attention Residual-Convolutional Neural Network model. The extracted features in the third step were selected by a new and simple algorithm (NCA-ReliefF Matched Index) based on matching the indexes obtained from the Neighborhood Component Analysis and Relief algorithms. In the last process, classification was done with the support vector machine algorithm. The proposed methodology was used for dual and multi-class classification tasks and evaluated on a dataset in the Kaggle database named Severstal: Steel Defect Detection. [ABSTRACT FROM AUTHOR]

The soil conditions substantially influence the intensity of underground metro train-induced ground vibration. However, assessing the influence of metro train-induced vibration on nearby structures is difficult due to problems in acquiring soil characteristics' field data. An optimized artificial neural network (ANN)-based hybrid prediction model is proposed to predict the underground metro train-induced ground vibration to address this issue. Particle swarm optimization (PSO) and genetic algorithm (GA) are used in the hybrid model to optimize the neural network's weights and biases. The datasets obtained by the validated numerical model are used to train the neural network. An explicit time domain, two-dimensional (2D) numerical model is developed using the finite element method (FEM) based on a two-step methodology and validated with experimental results of Delhi metro sites. Good agreement is observed between the numerical and experimental results in both time and frequency domains. The proposed hybrid PSO-ANN and GA-ANN model considered several soil properties such as density, Poisson's ratio, damping, Young's modulus and wave speed while predicting train-induced ground vibration. The PSO-ANN model predicted the outcome with an average error of 0.86%. Moreover, the proposed PSO-ANN model is also compared with the Federal Transit Administration's (FTA) semi-empirical approach of vibration assessment, linear support vector machine (SVM) and classification and regression tree (CART). The hybrid PSO-ANN model predicted the outcomes with greater accuracy than the basic ANN model, linear SVM, CART, GA-ANN and FTA's approach. Additionally, the hybrid model can consider the effect of parameters such as vehicle characteristics, suspension system, speed and tunnel depth. [ABSTRACT FROM AUTHOR]

When missing data are imputed by any method, there is some uncertainty associated with the imputed value. Consequently, when such imputed data are classified, some uncertainty will be propagated to the classifier output. This leads to two issues to address. First, reducing the uncertainty in the imputed value. Second, modeling and processing of the uncertainty associated with the classifier output to arrive at a better decision. To deal with the first issue, we use a latent space representation, while for the second issue we use Dempster-Shafer evidence theory. First, we train a neural network using the data without any missing value to generate a latent space representation of the input. The complete data set is now extended by deleting every feature once. These missing values are estimated using a nearest neighbor-based scheme. The network is then refined using this extended dataset to obtain a better latent space. This mechanism is expected to reduce the effect of the missing data on the latent space representation. Using the latent space representation of the complete data, we train two classifiers, support vector machines and evidential t-nearest neighbors. To classify an input with a missing value, we make a rough estimate of the missing value using the nearest neighbor rule and generate its latent space representation for classification by the classifiers. Using each classifier output, we generate a basic probability assignment (BPA) and all BPAs are combined to get an overall BPA. Final classification is done using Pignistic probabilities computed on the overall BPA. We use three different ways to defining BPAs. To avoid some problems of Dempster's rule of aggregation, we also use several alternative aggregations including some T-norm-based methods. Note that, T-norm has been used for combination of belief function in Pichon and Denœux (in: NAFIPS 2008: 2008 annual meeting of the North American fuzzy information processing society, pp 1–6, 2008). To demonstrate the superiority of the proposed method, we compare its performance with four state-of-the-art techniques using both artificial and real datasets. [ABSTRACT FROM AUTHOR]

One of the important types of lightweight concrete is expanded polystyrene (EPS) concrete which can be constructed by substituting part of coarse aggregates of the concrete with EPS beads. The characteristics of EPS concrete is greatly dependent on the material which is used in its manufacturing. Present study aims to develop two soft computing approaches according to radial basis function neural network (RBFNN) and coupled simulated annealing-least square support vector machine (CSA-LSSVM) models and a regression model for approximation the compressive strength of various types of concrete with main focus of EPS concrete. The regression model was considered as the benchmark model and the outcomes of RBFNN and CSA-LSSVM models were compared with it. The results revealed that the developed CSA-LSSVM and RBFNN approaches can be efficiently utilized to estimate the compressive strength of various kinds of concrete. The outcomes of the regression model were not as accurate as those of the machine learning approaches. In addition, the CSA-LSSVM model developed by the use of radial basis kernel function was decided as a better model. The R2, AARD%, RMSE, and MAPE for the benchmark model were 0.7555, 8.87%, 25.49, and 0.00887; however, these values for CSA-LSSVM were obtained as 0.9970, 3.01%, 1.08, and 0.0301 and for the RBFNN model were obtained to be 0.9953, 4.84%, 1.23, and 0.0484, respectively. [ABSTRACT FROM AUTHOR]

In this study, the performance of a counter-flow Ranque–Hilsch vortex tube (RHVT) was investigated experimentally using working fluid nitrogen, and the thermal performance was modeled using different modeling methods with these experimental results. Each method was compared with the others. The variation of ΔT, which is the measure of temperature separation in RHVT, was investigated by using nozzle number, the thermal conductivity of nozzle material, inlet pressure, specific heat, and density of working fluid. In the study, the prediction models of linear, k-nearest neighbor (kNN), random forest (RF), and support vector machine (SVM) regression were trained with measured thermal performance using a specific portion of the experimental data and tested with the remaining data. Two different train and test datasets were used with the ratio of experimental data as 90–10% and 80–20%, respectively. The highest accuracy ratio was determined at the end of the four methods with SVM regression as 96.01% when using the train and test datasets as 90–10%, respectively. The percent accuracy of the other models under the same conditions was calculated as 95.7, 90.87, and 78.36% for RF, kNN, and linear, respectively. [ABSTRACT FROM AUTHOR]

Specific language impairment (SLI) is one of the most common diseases in children, and early diagnosis can help to obtain better timely therapy economically. It is difficult and time-consuming for clinicians to accurately detect SLI through standard clinical assessments. Hence, machine learning algorithms have been developed to assist in the accurate diagnosis of SLI. This work aims to investigate the graph of the favipiravir molecule-based feature extraction function and propose an accurate SLI detection model using vowels. We proposed a novel handcrafted machine learning framework. This architecture comprises the favipiravir molecular structure pattern, statistical feature extractor, wavelet packet decomposition (WPD), iterative neighborhood component analysis (INCA), and support vector machine (SVM) classifier. Two feature extraction models, statistical and textural, are employed in the handcrafted feature generation methodology. A new nature-inspired graph-based feature extractor that uses the chemical depiction of the favipiravir (favipiravir became popular with the COVID-19 pandemic) is employed for feature extraction. Finally, the proposed favipiravir pattern, statistical feature extractor, and wavelet packet decomposition are used to create a feature vector. Moreover, a statistical feature extractor is used in this work. The WPD generates multilevel features, and the most meaningful features are selected using the NCA feature selector. Finally, these chosen features are fed to SVM classifier for automated classification. Two validation methods, (i) leave one subject out (LOSO) and (ii) tenfold cross-validations (CV), are used to obtain robust classification results. Our proposed favipiravir pattern-based model developed using a vowel dataset can detect SLI children with an accuracy of 99.87% and 98.86% using tenfold and LOSO CV strategies, respectively. These results demonstrated the high vowel classification ability of the proposed favipiravir pattern-based model. [ABSTRACT FROM AUTHOR]

Credit card churn prediction, insurance fraud detection, and loan default prediction are all critical analytical customer relationship management (ACRM) problems. Since these events occur infrequently, datasets for these problems are highly unbalanced. Consequently, when trained on such unbalanced datasets, all machine learning classifiers tend to produce high false positive rates. We propose two methods for data balancing. To oversample the minority class, we proposed an innovative GAN called chaoticGAN, where we employed chaotic noise as input for the generator. We also employed the traditional GAN (Goodfellow et al. in Adv Neural Inf Process Syst, 2014. https://doi.org/10.1145/3422622), Wasserstein GAN (Arjovsky et al. in Wasserstein GAN, 2017. https://arxiv.org/abs/1701.07875), and CTGAN (Xu et al. in Modeling Tabular Data using Conditional GAN. https://arxiv.org/pdf/1907.00503) independently for baseline comparison. On the data balanced by GANs, we employed a host of machine learning classifiers, including Random Forest, Decision Tree, Support Vector Machine (SVM), Logistic Regression (LR), multi-layer perceptron (MLP) and Light gradient boosting machine (LGBM) to demonstrate the efficacy of our approaches. In the second approach, we augment the oversampled synthetic minority class data obtained by GAN and its variants with the undersampled majority class data obtained by one class support vector machine (OCSVM) (Tax et al. in Mach Learn 54:45–66, 2014). We passed the entire modified dataset to build the classifiers. Our proposed approaches outperform earlier studies on the same datasets in terms of the area under the ROC curve (AUC). Further, our proposed chaoticGAN and its hybrid turned out to be statistically similar to the state-of-the-art CTGAN on all datasets while being significant over other methods w.r.t AUC over tenfold cross-validation. [ABSTRACT FROM AUTHOR]

The least squares support vector machine (LS-SVM) is an effective method to deal with classification and regression problems and has been widely studied and applied in the fields of machine learning and pattern recognition. The learning algorithms of the LS-SVM are usually conjugate gradient (CG) and sequential minimal optimization (SMO) algorithms. Based on this, we propose a conjugate functional gain SMO algorithm and theoretically prove its asymptotic convergence. This algorithm combines the conjugate direction method and the functional gain SMO algorithm with second-order information, which increases the functional gain of the plain SMO algorithm. In addition, we also provide a generalized SMO-type algorithm framework with a simple iterative format and easy implementation for other LS-SVM training algorithms. The numerical results show that the execution time of this algorithm is significantly shorter than that of the other plain SMO-type algorithms and CG-type algorithms. [ABSTRACT FROM AUTHOR]

Non-motor manifestations of Parkinson's disease (PD) appear early and have a significant impact on the quality of life of patients, but few studies have evaluated their predictive potential with machine learning algorithms. We evaluated 9 algorithms for discriminating PD patients from controls using a wide collection of non-motor clinical PD features from two databases: Biocruces (96 subjects) and PPMI (687 subjects). In addition, we evaluated whether the combination of both databases could improve the individual results. For each database 2 versions with different granularity were created and a feature selection process was performed. We observed that most of the algorithms were able to detect PD patients with high accuracy (>80%). Support Vector Machine and Multi-Layer Perceptron obtained the best performance, with an accuracy of 86.3% and 84.7%, respectively. Likewise, feature selection led to a significant reduction in the number of variables and to better performance. Besides, the enrichment of Biocruces database with data from PPMI moderately benefited the performance of the classification algorithms, especially the recall and to a lesser extent the accuracy, while the precision worsened slightly. The use of interpretable rules obtained by the RIPPER algorithm showed that simply using two variables (autonomic manifestations and olfactory dysfunction), it was possible to achieve an accuracy of 84.4%. Our study demonstrates that the analysis of non-motor parameters of PD through machine learning techniques can detect PD patients with high accuracy and recall, and allows us to select the most discriminative non-motor variables to create potential tools for PD screening. [ABSTRACT FROM AUTHOR]

Verifying the genuineness of official documents, such as bank checks, certificates, contract forms, bonds, etc., remains a challenging task when it comes to accuracy and robustness. Here, the genuineness is related to the degree of match of the signature contained in the documents relating to the original signatures of the authorized person. Signatures of authorized persons are considered known in advance. In this paper, a novel feature set is introduced based on quasi-straightness of boundary pixel runs for signature verification. We extract the quasi-straight line segments using elementary combinations of the directional codes from the signature boundary pixels and subsequently we obtain the feature set from various quasi-straight line classes. The quasi-straight line segments provide a blending of straightness and small curvatures resulting in a robust feature set for the verification of signatures. We have used Support Vector Machine (SVM) for classification and have shown results on standard signature datasets like CEDAR (Center of Excellence for Document Analysis and Recognition) and GPDS-100 (Grupo de Procesado Digital de la Senal). The results establish how the proposed method outperforms the existing state of the art. [ABSTRACT FROM AUTHOR]

Keyword extraction is one of the main problems in clustering and linking textual content. In literature, several machine learning approaches were proposed for keyword and keyphrase extraction. However, the state-of-the-art performance results are still below the expectations. In this paper, we propose a novel hybrid keyword extraction model, HybridKEM. The proposed model addresses the keyword extraction problem as a sequence labelling task. Naive Bayes (NB), Polynomial Regression (PR) Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), and Random Forest (RF) classification algorithms were trained separately in the Token Classification module of the model. The Token Classification process was performed by using text, graphic, embedding, and set features in the model. The performance of the model was evaluated using the Inspec, Semeval-2017, 500N-KPCrowd datasets, which are widely used in studies in the literature, and two newly collected, TRDizinEn and DergiParkEn datasets. The model achieved an average F 1 -score of 0.664 for all datasets. The highest F 1 -score (0.74) was obtained with the TRDizinEn dataset. [ABSTRACT FROM AUTHOR]

Small sphere and large margin support vector machine (SSLM) is an effective method for imbalanced data classification. However, the hinge loss used in SSLM easily leads to sensitivity to the noises and thus yields poor generalization performance since the outliers gain the largest penalties. In this paper, we propose a Ramp loss small sphere and large margin support vector machine (Ramp SSLM) for imbalanced data classification to improve the performance of SSLM. In comparison with SSLM, our model can incorporate noises, has less support vectors and thus owns better scaling properties. The non-convexity of Ramp SSLM can be efficiently solved by the concave-convex procedure (CCCP), which contains a sequence of convex problems. Furthermore, a sequential minimal optimization (SMO) decomposition method is employed to deal with the large-scale datasets. Experiments on an artificial, ten benchmark datasets and Chinese wine dataset are conducted and evaluation metrics such as g-means and F 1 score are adopted. Our method achieves better performance than other state-of-the-art algorithms, which shows the stability and effectiveness of our proposed algorithm. [ABSTRACT FROM AUTHOR]

With the iterative update of computer technology, the penetration of computer and other Internet technologies in human–computer interaction systems has become more and more extensive, and the human–computer interaction methods have quietly undergone huge changes. Gesture recognition has gradually become a hot spot in the field of human–computer interaction now, which has a wide range of application prospects and research value. The color segmentation experiment shows that the skin color of the gesture in the YCrCg space has better clustering properties than in the YCrCb space. In the preprocessing of gesture images, an improved Otsu method is proposed to improve the real-time performance to realize the threshold segmentation of the human hand; then the morphological processing is carried out, and the median filter method is used to achieve image denoising to improve image quality. A gesture recognition algorithm is designed: First, use Graph-SAGE to recognize the graph-structured data of the gesture, and then use the Adaboost algorithm to combine the two strong classifiers of the random forest and the support vector machine into a cascade classifier through the cascade structure. The output information of Graph-SAGE is classified and the meaning of the gesture is analyzed. On the test set, the average detection accuracy of the algorithm is 91.70%, the recall rate is 94.23%, and the average detection time per frame is 330 ms. [ABSTRACT FROM AUTHOR]