Extreme learning machines: a survey

Improved deep canonical correlation fusion approach for detection of early mild cognitive impairment

Detection of early mild cognitive impairment (EMCI) is clinically challenging as it involves subtle alterations in multiple brain sub-anatomic regions. Among different brain regions, the corpus callosum and lateral ventricles are primarily affected due to EMCI. In this study, an improved deep canonical correlation analysis (CCA) based framework is proposed to fuse magnetic resonance (MR) image features from lateral ventricular and corpus callosal structures for the detection of EMCI condition. For this, obtained structural MR images of healthy controls and EMCI subjects are preprocessed. Lateral ventricles and corpus callosum structures are segmented from these images and features are extracted. Extracted features from different brain structures are fused using non-linear orthogonal iteration-based deep CCA. Fused features are employed to differentiate healthy controls and EMCI condition using extreme learning machine classifier. Results indicate that fused callosal and ventricular features are able to detect EMCI. Improved deep CCA algorithm with tuned hyperparameters achieves the highest classifier performance with an F-score of 82.15%. The proposed framework is compared with state-of-the-art CCA approaches, and the results demonstrate its improved performance in EMCI detection. This highlights the potential of the proposed framework in the automated diagnosis of preclinical MCI conditions.

Traffic flow prediction based on improved deep extreme learning machine

A new hybrid prediction model is proposed for short-term traffic flow, which is based on Deep Extreme Learning Machine improved by Sparrow Search Algorithm (SSA-DELM). Firstly, Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise algorithm (ICEEMDAN) is employed to improve prediction accuracy. Then multiple Intrinsic Mode Function components (IMF) can be obtained. Secondly, Permutation Entropy algorithm (PE) is used to analyze the randomness of IMFs. Finally, different prediction models can be built according to the randomness characteristics. SSA-DELM prediction models are established for IMFs with large permutation entropy values. The IMFs with small permutation entropy values are put into ARIMA prediction models. To obtain the predicted traffic flow, different IMFs predicted values are added together. Two actual signalized intersections are selected to verify the performance of the new proposed model in this paper. Several prediction models based on different algorithms are built. The results obtained by MATLAB software show that the prediction errors of the new proposed model are the smallest and the fitting effect with the measured data is the best, which can effectively improve prediction accuracy.

Fibrosis and inflammatory activity diagnosis of chronic hepatitis C based on extreme learning machine

The traditional diagnosis of chronic hepatitis C usually relies on liver biopsy. Diagnosing chronic hepatitis C based on serum indices provides a non-invasive way to determine the stage of chronic hepatitis C without liver biopsy. In this paper, we proposed two automatic diagnosis systems for non-invasive diagnosis of chronic hepatitis C based on serum indices, an extreme learning machine (ELM) based auto-diagnosis method and a hybrid method using k-means clustering and ELM. The two proposed systems were used to predict the fibrosis stage and inflammatory activity grade of patients with chronic hepatitis C by analyzing their serum index observations. ELM has superiorities such as simple structure and fast calculation speed and can provide good diagnosis performance. To overcome the problem of class-imbalance, outliers and small sample size, we also proposed a method hybridizing k-means and ELM. It employed the k-means clustering to generate new robust training samples and then employed the new generated training samples to train an ELM for chronic hepatitis C diagnosis. The proposed methods were tested on 123 real clinical cases. Experimental results show that the proposed methods outperform the state-of-the-art methods for the fibrosis stage and inflammatory activity grade diagnosis tasks.

Development of a novel modeling framework based on weighted kernel extreme learning machine and ridge regression for streamflow forecasting

A precise streamflow forecast is crucial in hydrology for flood alerts, water quantity and quality management, and disaster preparedness. Machine learning (ML) techniques are commonly employed for hydrological prediction; however, they still face certain drawbacks, such as the need to optimize the appropriate predictors, the ability of the models to generalize across different time horizons, and the analysis of high-dimensional time series. This research aims to address these specific drawbacks by developing a novel framework for streamflow forecasting. Specifically, a hybrid ML model, WKELM-R, is developed to predict streamflow based on daily discharge and precipitation. The model combines ridge regression (RR), locally weighted linear regression (LWLR), and kernel extreme learning machine (KELM) to enhance multi-step-ahead predictions by accounting for both linear and nonlinear characteristics. In data preprocessing, this study applies multivariate variational mode decomposition (MVMD) for decomposition to handle non-stationarity and complexity, Boruta-XGBoost for feature selection to select the optimal inputs and decrease the dimension, and gradient-based optimizer (GBO) for adjustment of model parameters to overcome the need to optimize the appropriate predictors. To demonstrate the ability to handle real-world conditions and different time horizons, WKELM-R was applied to a watershed in North Dakota, USA to forecast discharge for three different time horizons. The results were compared with those from the existing standalone and hybrid models by multi-criteria decision-making (MCDM), demonstrating the efficacy and unique capabilities of the new hybrid model in streamflow forecasting (for the testing level at t + 3: R = 0.992, RMSE = 0.426, NSE = 0.983; at t + 7: R = 0.997, RMSE = 0.249, NSE = 0.994; at t + 14: R = 0.996, RMSE = 0.304, NSE = 0.991).

Retinal imaging based glaucoma detection using modified pelican optimization based extreme learning machine

Glaucoma is defined as progressive optic neuropathy that damages the structural appearance of the optic nerve head and is characterized by permanent blindness. For mass fundus image-based glaucoma classification, an improved automated computer-aided diagnosis (CAD) model performing binary classification (glaucoma or healthy), allowing ophthalmologists to detect glaucoma disease correctly in less computational time. We proposed learning technique called fast discrete curvelet transform with wrapping (FDCT-WRP) to create feature set. This method is entitled extracting curve-like features and creating a feature set. The combined feature reduction techniques named as principal component analysis and linear discriminant analysis, have been applied to generate prominent features and decrease the feature vector dimension. Lastly, a newly improved learning algorithm encompasses a modified pelican optimization algorithm (MOD-POA) and an extreme learning machine (ELM) for classification tasks. In this MOD-POA+ELM algorithm, the modified pelican optimization algorithm (MOD-POA) has been utilized to optimize the parameters of ELM's hidden neurons. The effectiveness has been evaluated using two standard datasets called G1020 and ORIGA with the [Formula: see text]-fold stratified cross-validation technique to ensure reliable evaluation. Our employed scheme achieved the best results for both datasets obtaining accuracy of 93.25% (G1020 dataset) and 96.75% (ORIGA dataset), respectively. Furthermore, we have utilized seven Explainable AI methodologies: Vanilla Gradients (VG), Guided Backpropagation (GBP ), Integrated Gradients ( IG), Guided Integrated Gradients (GIG), SmoothGrad, Gradient-weighted Class Activation Mapping (GCAM), and Guided Grad-CAM (GGCAM) for interpretability examination, aiding in the advancement of dependable and credible automation of healthcare detection of glaucoma.

Rapid quantification of single component oil in perilla oil blends by ultraviolet-visible spectroscopy combined with chemometrics

As a unconventional oil, perilla oil is much more expensive than conventional oils since it has the highest content of α-linolenic acid among vegetable oils. Thus the adulteration of perilla oil is serious, which needs to be solved. In this study, the single component oil in perilla oil blends were first quantitatively analyzed by ultraviolet-visible (UV-vis) spectroscopy combined with chemometric methods. Soybean oil and palm oil were added into perilla oil to form binary and ternary perilla oil blends. Partial least squares (PLS), back propagation-artificial neural network (BP-ANN), support vector regression (SVR) and extreme learning machine (ELM) were compared and the best model was selected for calibration. In order to improve the prediction performance of the calibration model, ten preprocessing methods and five variable selection methods were investigated. Results show that PLS was the best calibration method for binary and ternary perilla oil blends. For binary perilla oil blends, the correlation coefficients of prediction (Rp) obtained by PLS were both above 0.99, which does not need preprocessing and variable selection. For ternary perilla oil blends, after the best continuous wavelet transform (CWT) preprocessing and discretized whale optimization algorithm (WOA) variable selection, the Rp values obtained by the best model CWT-WOA-PLS were all above 0.97. This research provides a common framework for calibration of perilla oil blends, which maybe a promising method for quality control of perilla oil in industry.

A lightweight method of integrated local load forecasting and control of edge computing in active distribution networks

The strong resource constraints of edge-computing devices and the dynamic evolution of load characteristics put forward higher requirements for forecasting methods of active distribution networks. This paper proposes a lightweight adaptive ensemble learning method for local load forecasting and predictive control of active distribution networks based on edge computing in resource constrained scenarios. First, the adaptive sparse integration method is proposed to reduce the model scale. Then, the auto-encoder is introduced to downscale the model variables to further reduce computation time and storage overhead. An adaptive correction method is proposed to maintain the adaptability. Finally, a multi-timescale predictive control method for the edge side is established, which realizes the collaboration of local load forecasting and control. All cases can be deployed on an actual edge-computing device. Compared to other benchmark methods and the existing researches, the proposed method can minimize the model complexity without reducing the forecasting accuracy.

Integrated metaheuristic algorithms with extreme learning machine models for river streamflow prediction

Accurate river streamflow prediction is pivotal for effective resource planning and flood risk management. Traditional river streamflow forecasting models encounter challenges such as nonlinearity, stochastic behavior, and convergence reliability. To overcome these, we introduce novel hybrid models that combine extreme learning machines (ELM) with cutting-edge mathematical inspired metaheuristic optimization algorithms, including Pareto-like sequential sampling (PSS), weighted mean of vectors (INFO), and the Runge-Kutta optimizer (RUN). Our comparative assessment includes 20 hybrid models across eight metaheuristic categories, using streamflow data from the Aswan High Dam on the Nile River. Our findings highlight the superior performance of mathematically based models, which demonstrate enhanced predictive accuracy, robust convergence, and sustained stability. Specifically, the PSS-ELM model achieves superior performance with a root mean square error of 2.0667, a Pearson's correlation index (R) of 0.9374, and a Nash-Sutcliffe efficiency (NSE) of 0.8642. Additionally, INFO-ELM and RUN-ELM models exhibit robust convergence with mean absolute percentage errors of 15.21% and 15.28% respectively, a mean absolute errors of 1.2145 and 1.2105, and high Kling-Gupta efficiencies values of 0.9113 and 0.9124, respectively. These findings suggest that the adoption of our proposed models significantly enhances water management strategies and reduces any risks.

Robust face recognition using quaternion interval type II fuzzy logic-based feature extraction on colour images

In this paper, we propose a new robust and fast learning technique by investigating the effect of integration of quaternion and interval type II fuzzy logic along with non-iterative, parameter free deterministic learning machine (DLM) pertaining to face recognition problem. The traditional learning techniques did not account colour information and degree of pixel wise association of individual pixel of a colour face image in their network. Therefore, this paper presents a new technique named quaternion interval type II based deterministic learning machine (QIntTyII-DLM), which considers the interrelationship between three colour channels viz. red, green, and blue (RGB) by representing each colour pixel of a colour image in quaternion number sequence. Here, quaternion vector representation of a colour face image is fuzzified using interval type II fuzzy logic. This reduces the redundancy between pixels of different colour channels and also transforms colour channels of the image to orthogonal colour space. Thereafter, classification is performed using DLM. Experiments performed (on four standard datasets AR, Georgia Tech, Indian, face (female) and faces 94 (male) face datasets) and comparison done with other existing techniques proves that the proposed technique gives better results in terms of percentage error rate (reduces approximately 10-12%) and computational speed.

Bidirectional machine learning-assisted sensitivity-based stochastic searching approach for groundwater DNAPL source characterization

In this study, we designed a machine learning-based parallel global searching method using the Bayesian inversion framework for efficient identification of dense non-aqueous phase liquid (DNAPL) source characteristics and contaminant transport parameters in groundwater. Swarm intelligence organized hybrid-kernel extreme learning machine (SIO-HKELM) was proposed to approximate the forward and inverse input-output correlation with a high accuracy using the DNAPL transport numerical simulation model. An adaptive inverse-HKELM was established for preliminary estimation of the source characteristics and contaminant transport parameters to correct prior information and generate high-quality initial starting points of parallel searching. A local accurate forward-HKELM surrogate of the numerical model was embedded in the searching system for avoiding repetitive CPU-demanding likelihood evaluations. A sensitivity-based Metropolis criterion (MC), incorporating the dynamic particle swarm optimization (SD-PSO) algorithm, was developed for improving the search ergodicity and realizing precise inversion of all the unknown variables with drastic variations in sensitivity to the likelihood function. Results showed that the generalization capability and robustness of SIO-HKELM were superior to those of the traditional machine learning methods, including KELM and support vector regression (SVR), and it sufficiently approximated the forward and inverse input-output mapping of the numerical model with testing determination coefficients of 0.9944 and 0.6440, respectively. With high-quality prior information and initial starting points generated by the adaptive inverse-HKELM feed approach, the uncertainty in the inversion outputs was reduced, and the searching process rapidly converged to reasonable posterior distributions in around 60 iterations. Compared with the widely used multichain Markov chain Monte Carlo (MCMC) approach, the parallel searching lines generated by SD-PSO-MC adequately covered the searching space, and the "equifinality" effect was more effectively restrained by reducing the relative errors of all the point estimations to less than 8%. Therefore, the real source information reflected by the statistical characteristics of the SD-PSO-MC inversion outputs was more precise than that obtained using the multichain MCMC approach.

Experimental Property Reconstruction in a Photonic Quantum Extreme Learning Machine

Recent developments have led to the possibility of embedding machine learning tools into experimental platforms to address key problems, including the characterization of the properties of quantum states. Leveraging on this, we implement a quantum extreme learning machine in a photonic platform to achieve resource-efficient and accurate characterization of the polarization state of a photon. The underlying reservoir dynamics through which such input state evolves is implemented using the coined quantum walk of high-dimensional photonic orbital angular momentum and performing projective measurements over a fixed basis. We demonstrate how the reconstruction of an unknown polarization state does not need a careful characterization of the measurement apparatus and is robust to experimental imperfections, thus representing a promising route for resource-economic state characterization.

Research on Circuit Breaker Operating Mechanism Fault Diagnosis Method Combining Global-Local Feature Extraction and KELM

In response to the lack of generality in feature extraction using modal decomposition methods and the susceptibility of diagnostic performance to parameter selection in traditional mechanical fault diagnosis of high-voltage circuit breaker operating mechanisms, this paper proposes a Global-Local feature extraction method based on Generalized S-Transform (S-Translate) combined with Gray Level Co-Occurrence Matrix (GLCM) and complemented by Maximum Relevance and Minimum Redundancy (mRMR) feature selection. The GL (Global-Local)-mRMR-KELM fault diagnosis model is proposed, which employs the Kernel Extreme Learning Machine (KELM). In this model, the original time-frequency domain features and the time-frequency features of the Generalized S-Transform matrix of vibration signals under different states of the circuit breaker are first extracted as global features. Then, the GLCM is obtained to extract texture features as local features. Finally, the mRMR and KELM are comprehensively applied to perform feature selection and classification on the dataset, thereby accomplishing the fault diagnosis of the circuit breaker's operating mechanism. In this study, the 72.5 kV SF6 circuit breaker operating mechanism is taken as the research object, and three types of mechanical faults are simulated to obtain a vibration signal. Experimental results verify the effectiveness of the proposed GL-mRMR-KELM model, achieving a diagnostic accuracy of 96%. This research provides a feasible approach for the fault diagnosis of circuit breaker operating mechanisms.

Artificial intelligence-based framework for precise prediction of asphaltene particle aggregation kinetics in petroleum recovery

The precipitation and deposition of asphaltene on solid surfaces present a significant challenge throughout all stages of petroleum recovery, from hydrocarbon reservoirs in porous media to wellbore and transfer pipelines. A comprehensive understanding of asphaltene aggregation phenomena is crucial for controlling deposition issues. In addition to experimental studies, accurate prediction of asphaltene aggregation kinetics, which has received less attention in previous research, is essential. This study proposes an artificial intelligence-based framework for precisely predicting asphaltene particle aggregation kinetics. Different techniques were utilized to predict the asphaltene aggregate diameter as a function of pressure, temperature, oil specific gravity, and oil asphaltene content. These methods included the adaptive neuro-fuzzy interference system (ANFIS), radial basis function (RBF) neural network optimized with the Grey Wolf Optimizer (GWO) algorithm, extreme learning machine (ELM), and multi-layer perceptron (MLP) coupled with Bayesian Regularization (BR), Levenberg-Marquardt (LM), and Scaled Conjugate Gradient (SCG) algorithms. The models were constructed using a series of published data. The results indicate the excellent correlation between predicted and experimental values using various models. However, the GWO-RBF modeling strategy demonstrated the highest accuracy among the developed models, with a determination coefficient, average absolute relative deviation percent, and root mean square error (RMSE) of 0.9993, 1.1326%, and 0.0537, respectively, for the total data.

Motor imagery classification using sparse representations: an exploratory study

The non-stationary nature of the EEG signal poses challenges for the classification of motor imagery. sparse representation classification (SRC) appears as an alternative for classification of untrained conditions and, therefore, useful in motor imagery. Empirical mode decomposition (EMD) deals with signals of this nature and appears at the rear of the classification, supporting the generation of features. In this work we evaluate the combination of these methods in a multiclass classification problem, comparing them with a conventional method in order to determine if their performance is regular. For comparison with SRC we use multilayer perceptron (MLP). We also evaluated a hybrid approach for classification of sparse representations with MLP (RSMLP). For comparison with EMD we used filtering by frequency bands. Feature selection methods were used to select the most significant ones, specifically Random Forest and Particle Swarm Optimization. Finally, we used data augmentation to get a more voluminous base. Regarding the first dataset, we observed that the classifiers that use sparse representation have results equivalent to each other, but they outperform the conventional MLP model. SRC and SRMLP achieve an average accuracy of [Formula: see text] and [Formula: see text] respectively while the MLP is [Formula: see text], representing a gain between [Formula: see text] and [Formula: see text]. The use of EMD in relation to other feature processing techniques is not superior. However, EMD does not influence negatively, there is an opportunity for improvement. Finally, the use of data augmentation proved to be important to obtain relevant results. In the second dataset, we did not observe the same results. Models based on sparse representation (SRC, SRMLP, etc.) have on average a performance close to other conventional models, but without surpassing them. The best sparse models achieve an average accuracy of [Formula: see text] among the subjects in the base, while other model reach [Formula: see text]. The improvement of self-adaptive mechanisms that respond efficiently to the user's context is a good way to achieve improvements in motor imagery applications. However, other scenarios should be investigated, since the advantage of these methods was not proven in all datasets studied. There is still room for improvement, such as optimizing the dictionary of sparse representation in the context of motor imagery. Investing efforts in synthetically increasing the training base has also proved important to reduce the costs of this group of applications.

Deep extreme learning machine with knowledge augmentation for EEG seizure signal recognition

Introduction

Intelligent recognition of electroencephalogram (EEG) signals can remarkably improve the accuracy of epileptic seizure prediction, which is essential for epileptic diagnosis. Extreme learning machine (ELM) has been applied to EEG signals recognition, however, the artifacts and noises in EEG signals have a serious effect on recognition efficiency. Deep learning is capable of noise resistance, contributing to removing the noise in raw EEG signals. But traditional deep networks suffer from time-consuming training and slow convergence.

Methods

Therefore, a novel deep learning based ELM (denoted as DELM) motivated by stacking generalization principle is proposed in this paper. Deep extreme learning machine (DELM) is a hierarchical network composed of several independent ELM modules. Augmented EEG knowledge is taken as complementary component, which will then be mapped into next module. This learning process is so simple and fast, meanwhile, it can excavate the implicit knowledge in raw data to a greater extent. Additionally, the proposed method is operated in a single-direction manner, so there is no need to perform parameters fine-tuning, which saves the expense of time.

Results

Extensive experiments are conducted on the public Bonn EEG dataset. The experimental results demonstrate that compared with the commonly-used seizure prediction methods, the proposed DELM wins the best average accuracies in 13 out of the 22 data and the best average F-measure scores in 10 out of the 22 data. And the running time of DELM is more than two times quickly than deep learning methods.

Discussion

Therefore, DELM is superior to traditional and some state-of-the-art machine learning methods. The proposed architecture demonstrates its feasibility and superiority in epileptic EEG signal recognition. The proposed less computationally intensive deep classifier enables faster seizure onset detection, which is showing great potential on the application of real-time EEG signal classification.

Extreme learning machine for soil temperature prediction using only air temperature as input

Soil temperature (TS) is a crucial parameter in many fields, especially agriculture. In developing countries like Algeria, the soil temperatures (TS) and the meteorological data are limited. This study investigates the use of Extreme Learning Machine (ELM) for the accurate prediction of daily ST at three different depths (30 cm, 60 cm, and 100 cm) using a minimal number of climatic inputs. The inputs used in this study include maximum and minimum air temperatures, relative humidity, and day of the year (DOY) as a representative of the temporal component. Five different combinations of inputs were used to develop ELM models and determine the best set of input variables. The ELM models were then compared with traditional methods such as multiple linear regression, artificial neural networks, and adaptive neuro-fuzzy inference system. Based on evaluation metrics such as R, RMSE, and MAPE, the ELM models with air temperatures and DOY as inputs (ELM-M0 and ELM-M3) demonstrated superior performance at all depths when compared to the other techniques. The most accurate predictions were found at a depth of 100 cm using the ELM-M3 model, which employed inputs of minimum and maximum air temperatures and DOY, with R value of 0.98, RMSE of 0.68 °C, and MAPE of 3.4%. The results demonstrate that the inclusion of DOY in the climatic dataset significantly enhances the performance and accuracy of machine learning models for ST prediction. The ELM was found to be a fast, simple, effective, and useful tool for TS prediction.

Actuator fault detection and isolation on multi-rotor UAV using extreme learning neuro-fuzzy systems

Undetected partial actuator faults on multi-rotor UAVs can lead to system failures and uncontrolled crashes, necessitating the development of accurate and efficient fault detection and isolation (FDI) strategy. This paper proposes a hybrid FDI model for a quadrotor UAV that integrates an extreme learning neuro-fuzzy algorithm with a model-based extended Kalman filter (EKF). Three FDI models using Fuzzy-ELM, R-EL-ANFIS, and EL-ANFIS are compared based on training, validation performances, and sensitivity to weaker and shorter actuator faults. They are also tested online for linear and nonlinear incipient faults by measuring their isolation time delays and accuracies. The results show that the Fuzzy-ELM FDI model exhibits greater efficiency and sensitivity, while Fuzzy-ELM and R-EL-ANFIS FDI models demonstrate better performance than a conventional neuro-fuzzy algorithm, ANFIS.

A novel combined intelligent algorithm prediction model for the tunnel surface settlement

To ensure the safety and stability of the shield tunnel construction process, the ground settlement induced by the shield construction needs to be effectively predicted. In this paper, a prediction method combining empirical mode decomposition (EMD), chaotic adaptive sparrow search algorithm (CASSA), and extreme learning machine (ELM) is proposed. First, the EMD is used to decompose the settlement sequence into trend vectors and fluctuation vectors to fully extract the effective information of the sequence; Second, the sparrow search algorithm is improved by introducing Cubic chaotic mapping to initialize the population and adaptive factor to optimize the searcher's position formula, and the chaotic adaptive sparrow search algorithm is proposed; Finally, the CASSA-ELM prediction model is constructed by using CASSA to find the optimal values of weights and thresholds in the extreme learning machine. The fluctuation components and trend components decomposed by EMD are predicted one by one, and the prediction results are superimposed and reconstructed to obtain the predicted final settlement. Taking a shield interval in Jiangsu, China as an example, the meta-heuristic algorithm-optimized ELM model improves the prediction accuracy by 10.70% compared with the traditional ELM model. The combined EMD-CASSA-ELM prediction model can greatly improve the accuracy and speed of surface settlement prediction, and provide a new means for safety monitoring in shield tunnel construction. Intelligent prediction methods can predict surface subsidence more automatically and quickly, becoming a new development trend.

Online sequential, outlier robust, and parallel layer perceptron extreme learning machine models for sediment transport in sewer pipes

Sediment transport is a noteworthy task in the design and operation of sewer pipes. Decreasing sewer pipe hydraulic capacity and transport of pollution are the main consequences of continuous sedimentation. Among different design approaches, the non-deposition with deposited bed (NDB) method can be used for the design of large sewer pipes; however, existing models are established on limited data ranges and mostly applied conventional regression methods. The current study improves the NDB sediment transport modeling by utilizing wide data ranges, and furthermore, applying robust machine learning techniques. In the present study, the conventional extreme learning machine (ELM) technique and its advanced versions, namely the online sequential-extreme learning machine (OS-ELM), outlier robust-extreme learning machine (OR-ELM), and parallel layer perceptron-extreme learning machine (PLP-ELM) are used for the modeling. In the studies conducted in the literature, sediment deposited bed thickness (t_s) or deposited bed width (W_b) was used in the model structure as a deposited sediment variable, and therefore, different parameters in terms of t_s and W_b can be incorporated into the model structure. However, an uncertainty arises in the selection of the appropriate parameter among W_b/Y, t_s/Y, W_b/D, and t_s/D (Y is flow depth and D circular pipe diameter). In order to define the most appropriate parameter to best describe the impact of deposited sediment at the channel bottom in the modeling procedure, four various scenarios using four different parameters that incorporate deposited sediment variables at their structures as W_b/Y, t_s/Y, W/D, and t_s/D are considered for model development. It is found that models that incorporate sediment bed thickness (t_s) provide better results than those which use deposited bed width (W_b) in their structures. Among four different scenarios, models that utilized t_s/D dimensionless parameter, give superior results in contrast to their alternatives. Based on the outcomes, the OR-ELM approach outperformed ELM, OS-ELM, and PLP-ELM techniques. The results obtained from applied methods are compared to their corresponding models in the literature, indicating the superiority of the OR-ELM model. It is figured out that the thickness of the deposited bed is an effective variable in modeling NDB sediment transport in sewer pipes.

Prediction of line heating deformation on sheet metal based on an ISSA-ELM model

A prediction method based on an improved salp swarm algorithm (ISSA) and extreme learning machine (ELM) was proposed to improve line heating and forming. First, a three-dimensional transient numerical simulation of line heating and forming was carried out by applying a finite element simulation, and the influence of machining parameters on deformation was studied. Second, a prediction model for the ELM network was established based on simulation data, and the deformation of hull plate was predicted by the training network. Additionally, swarm intelligence optimization, particle swarm optimization (PSO), the seagull optimization algorithm (SOA), and the salp swarm algorithm (SSA) were studied while considering the shortcomings of the ELM, and the ISSA was proposed. Input weights and hidden layer biases of the ELM model were optimized to increase the stability of prediction results from the PSO, SOA, SSA and ISSA approaches. Finally, it was shown that the prediction effect of the ISSA-ELM model was superior by comparing and analyzing the prediction effect of each prediction model for line heating and forming.

Output Layer Structure Optimization for Weighted Regularized Extreme Learning Machine Based on Binary Method

In this paper, we focus on the redesign of the output layer for the weighted regularized extreme learning machine (WRELM). For multi-classification problems, the conventional method of the output layer setting, named “one-hot method”, is as follows: Let the class of samples be r; then, the output layer node number is r and the ideal output of s-th class is denoted by the s-th unit vector in Rr (1≤s≤r). Here, in this article, we propose a “binarymethod” to optimize the output layer structure: Let 2p−1<r≤2p, where p≥2, and p output nodes are utilized and, simultaneously, the ideal outputs are encoded in binary numbers. In this paper, the binary method is employed in WRELM. The weights are updated through iterative calculation, which is the most important process in general neural networks. While in the extreme learning machine, the weight matrix is calculated in least square method. That is, the coefficient matrix of the linear equations we solved is symmetric. For WRELM, we continue this idea. And the main part of the weight-solving process is a symmetry matrix. Compared with the one-hot method, the binary method requires fewer output layer nodes, especially when the number of sample categories is high. Thus, some memory space can be saved when storing data. In addition, the number of weights connecting the hidden and the output layer will also be greatly reduced, which will directly reduce the calculation time in the process of training the network. Numerical experiments are conducted to prove that compared with the one-hot method, the binary method can reduce the output nodes and hidden-output weights without damaging the learning precision.

COVID-19 Detection Mechanism in Vehicles Using a Deep Extreme Machine Learning Approach

COVID-19 is a rapidly spreading pandemic, and early detection is important to halting the spread of infection. Recently, the outbreak of this virus has severely affected people around the world with increasing death rates. The increased death rates are because of its spreading nature among people, mainly through physical interactions. Therefore, it is very important to control the spreading of the virus and detect people's symptoms during the initial stages so proper preventive measures can be taken in good time. In response to COVID-19, revolutionary automation such as deep learning, machine learning, image processing, and medical images such as chest radiography (CXR) and computed tomography (CT) have been developed in this environment. Currently, the coronavirus is identified via an RT-PCR test. Alternative solutions are required due to the lengthy moratorium period and the large number of false-negative estimations. To prevent the spreading of the virus, we propose the Vehicle-based COVID-19 Detection System to reveal the related symptoms of a person in the vehicles. Moreover, deep extreme machine learning is applied. The proposed system uses headaches, flu, fever, cough, chest pain, shortness of breath, tiredness, nasal congestion, diarrhea, breathing difficulty, and pneumonia. The symptoms are considered parameters to reveal the presence of COVID-19 in a person. Our proposed approach in Vehicles will make it easier for governments to perform COVID-19 tests timely in cities. Due to the ambiguous nature of symptoms in humans, we utilize fuzzy modeling for simulation. The suggested COVID-19 detection model achieved an accuracy of more than 90%.

A novel correlation Gaussian process regression-based extreme learning machine

An obvious defect of extreme learning machine (ELM) is that its prediction performance is sensitive to the random initialization of input-layer weights and hidden-layer biases. To make ELM insensitive to random initialization, GPRELM adopts the simple an effective strategy of integrating Gaussian process regression into ELM. However, there is a serious overfitting problem in kernel-based GPRELM (kGPRELM). In this paper, we investigate the theoretical reasons for the overfitting of kGPRELM and further propose a correlation-based GPRELM (cGPRELM), which uses a correlation coefficient to measure the similarity between two different hidden-layer output vectors. cGPRELM reduces the likelihood that the covariance matrix becomes an identity matrix when the number of hidden-layer nodes is increased, effectively controlling overfitting. Furthermore, cGPRELM works well for improper initialization intervals where ELM and kGPRELM fail to provide good predictions. The experimental results on real classification and regression data sets demonstrate the feasibility and superiority of cGPRELM, as it not only achieves better generalization performance but also has a lower computational complexity.

Meta-Heuristic Optimization Algorithm-Based Hierarchical Intrusion Detection System

Numerous network cyberattacks have been launched due to inherent weaknesses. Network intrusion detection is a crucial foundation of the cybersecurity field. Intrusion detection systems (IDSs) are a type of machine learning (ML) software proposed for making decisions without explicit programming and with little human intervention. Although ML-based IDS advancements have surpassed earlier methods, they still struggle to identify attack types with high detection rates (DR) and low false alarm rates (FAR). This paper proposes a meta-heuristic optimization algorithm-based hierarchical IDS to identify several types of attack and to secure the computing environment. The proposed approach comprises three stages: The first stage includes data preprocessing, feature selection, and the splitting of the dataset into multiple binary balanced datasets. In the second stage, two novel meta-heuristic optimization algorithms are introduced to optimize the hyperparameters of the extreme learning machine during the construction of multiple binary models to detect different attack types. These are combined in the last stage using an aggregated anomaly detection engine in a hierarchical structure on account of the model’s accuracy. We propose a software machine learning IDS that enables multi-class classification. It achieved scores of 98.93, 99.63, 99.19, 99.78, and 0.01, with 0.51 for average accuracy, DR, and FAR in the UNSW-NB15 and CICIDS2017 datasets, respectively.

An evolutionary machine learning for multiple myeloma using Runge Kutta Optimizer from multi characteristic indexes

Multiple myeloma (MM) is a malignant plasma cell disease that is the second most prevalent hematological malignancy in high-income nations and accounts for around 1.8% of all cancers and 18% of hematologic malignancies in the United States. In this research, we try to design a machine learning framework for MM diagnosis from multi characteristic indexes using slime mould Runge Kutta Optimizer (MSRUN) and kernel extreme learning machine, which is called as MSRUN-KELM. An efficient slime mould learning operator is introduced to the initial Runge Kutta Optimizer in MSRUN, ensuring that the trade-off between intensity and diversity is satisfied. The MSRUN was evaluated using IEEE CEC2014 benchmark functions, and the statistical results indicate a significant increase in the search performance of MSRUN. In MSRUN-KELM, kernel extreme machine learning is constructed on MM from multi-characteristic indexes with MSRUN, parameter optimization, and feature selection synchronized by MSRUN. The results of MSRUN-KELM on MM are accuracy of 93.88%, a Matthews correlation coefficient of 0.922677, and sensitivities of 93.41% and 93.19%. The suggested MSRUN-KELM may be utilized to analyze MM from multi-characteristic indexes well, and it can be treated as a potential tool for MM diagnosis.

Fault Diagnosis of Balancing Machine Based on ISSA-ELM

Balancing machine is a general equipment for dynamic balance verification of rotating parts, whether it breaks down or does not determine the accuracy of dynamic balance verification. In order to solve the problem of insufficient fault diagnosis accuracy of balancing machine, a fault diagnosis method of balancing machine based on the Improved Sparrow Search Algorithm (ISSA) optimized Extreme Learning Machine (ELM) was proposed. Firstly, iterative chaos mapping and Fuch chaos mapping were introduced to initialize the population and increase the population diversity. Secondly, the adaptive dynamic factor and Levy flight strategy were also introduced to update the individual positions and improve the model convergence speed. Finally, the fault feature vector was input to the ISSA-ELM model with the fault type as the output. The experiment showed that the fault diagnosis accuracy of ISSA-ELM is as high as 99.17%, which is 1.67%, 2.50%, 7.50%, and 17.50% higher than that of SSA-ELM, HHO-ELM, PSO-ELM, and ELM, respectively, further improving the prediction accuracy of the operation state of the balancing machine.

Robust Extreme Learning Machine Using New Activation and Loss Functions Based on M-Estimation for Regression and Classification

This paper provides an analysis of the combining effect of novel activation function and loss function based on M-estimation in application to extreme learning machine (ELM), a feed-forward neural network. Due to the computational efficiency and classification/prediction accuracy of ELM and its variants, they have been widely exploited in the development of new technologies and applications. However, in real applications, the performance of classical ELMs deteriorates in the presence of outliers, thus, negatively impacting the precision and accuracy of the system. To further enhance the performance of ELM and its variants, we proposed novel activation functions based on the psi function of M and redescend the M-estimation method along with the smooth $\mathcal{l}$ 2-norm weight-loss functions to reduce the negative impact of the outliers. The proposed psi functions of several M and redescending M-estimation methods are more flexible to make more distinct features space. For the first time, the idea of the psi function as an activation function in the neural network is introduced in the literature to ensure accurate prediction. In addition, new robust $\mathcal{l}$ 2 norm-loss functions based on M and redescending M-estimation are proposed to deal with outliers efficiently in ELM. To evaluate the performance of the proposed methodology against other state-of-the-art techniques, experiments have been performed in diverse environments, which show promising improvements in application to regression and classification problems.

Locally Weighted Principal Component Analysis-Based Multimode Modeling for Complex Distributed Parameter Systems

Global principal component analysis (PCA) has been successfully introduced for modeling distributed parameter systems (DPSs). In spite of the merits, this method is not feasible due to parameter variations and multiple operating domains. A novel multimode spatiotemporal modeling method based on the locally weighted PCA (LW-PCA) method is developed for large-scale highly nonlinear DPSs with parameter variations, by separating the original dataset into tractable subsets. This method implements the decomposition by making full use of the dependence among subset densities. First, the spatiotemporal snapshots are divided into multiple different Gaussian components by using a finite Gaussian mixture model (FGMM). Once the components are derived, a Bayesian inference strategy is then applied to calculate the posterior probabilities of each spatiotemporal snapshot belonging to each component, which will be regarded as the local weights of the LW-PCA method. Second, LW-PCA is adopted to calculate each locally weighted snapshot matrix, and the corresponding local spatial basis functions (SBFs) can be generated by the PCA method. Third, all the local temporal models are estimated using the extreme learning machine (ELM). Thus, the local spatiotemporal models can be produced with local SBFs and corresponding temporal model. Finally, the original system can be approximated using the sum form of each local spatiotemporal model. Unlike global PCA, which uses global SBFs to construct a global spatiotemporal model, LW-PCA approximates the original system by multiple local reduced SBFs. Numerical simulations verify the effectiveness of the developed multimode spatiotemporal model.

A Review of Advanced Methods for the Quantitative Analysis of Single Component Oil in Edible Oil Blends

Edible oil blends are composed of two or more edible oils in varying proportions, which can ensure nutritional balance compared to oils comprising a single component oil. In view of their economical and nutritional benefits, quantitative analysis of the component oils in edible oil blends is necessary to ensure the rights and interests of consumers and maintain fairness in the edible oil market. Chemometrics combined with modern analytical instruments has become a main analytical technology for the quantitative analysis of edible oil blends. This review summarizes the different oil blend design methods, instrumental techniques and chemometric methods for conducting single component oil quantification in edible oil blends. The aim is to classify and compare the existing analytical techniques to highlight suitable and promising determination methods in this field.

Fault Diagnosis of Coal Mill Based on Kernel Extreme Learning Machine with Variational Model Feature Extraction

Aiming at the typical faults in the coal mills operation process, the kernel extreme learning machine diagnosis model based on variational model feature extraction and kernel principal component analysis is offered. Firstly, the collected signals of vibration and loading force, corresponding to typical faults of coal mill, are decomposed by variational model decomposition, and the intrinsic model functions at different scales are obtained. Then, the eigenvectors consisting of feature energy and sample entropy in these functions are respectively calculated, and the kernel principal component analysis is used for noise removal and dimensionality reduction. Finally, the kernel extreme learning machine model is trained and tested with the dimension reduced feature vector as input and the corresponding coal mill state as output. The results show that the variational model decomposition extraction can improve the input features of the model compared with the single eigenvector model, and the kernel principal component analysis method can significantly reduce the information redundancy and the correlation of eigenvectors, which can effectively save time and cost, and improve the prediction performance of the model to some extent. The establishment of this model provides a new idea for the study of coal mill fault diagnosis.

Exploration of China's net CO2 emissions evolutionary pathways by 2060 in the context of carbon neutrality

In the context of global climate governance, as the biggest carbon emitter, China bears momentous responsibility for mitigating emissions. Especially after the carbon neutrality target is proposed, it is urgent for China to seek a feasible pathway to achieve net-zero carbon dioxide (CO₂) emissions by 2060. With the aims of exploring the net-zero emission pathways, an integrated prediction model incorporating the extreme learning machine (ELM) network, the Aquila optimizer (AO) technique, and the Elastic Net (EN) regression method is constructed. Then the prediction model is employed to project CO₂ emissions and forest carbon sinks during 2021-2060 under the nine designed scenarios. The simulation results reveal that China has the potential to achieve net-zero CO₂ emissions by 2060 under the combined effects of reducing emissions and increasing forest carbon sinks. Specifically, the total CO₂ emissions will be peaked at 11441 million tons CO₂ (MtCO₂) in 2029. The post-peak carbon reduction rate should be 8% per year, and the average annual forest carbon sink is required to be 209.45 TgC/year during 2021-2060. In addition, in accordance with the optimal carbon neutrality pathway, the GDP per capita growth rate should be maintained at 5.5% during the period of 2021-2030, China's urbanization rate should be increased to 72% in 2030, and the total energy consumption should be limited to a peak value of 6000 million tons of coal equivalent (Mtce) in 2030.

Extreme Learning Machine Design for Dealing with Unrepresentative Features

Extreme Learning Machines (ELMs) have become a popular tool for the classification of electroencephalography (EEG) signals for Brain Computer Interfaces. This is so mainly due to their very high training speed and generalization capabilities. Another important advantage is that they have only one hyperparameter that must be calibrated: the number of hidden nodes. While most traditional approaches dictate that this parameter should be chosen smaller than the number of available training examples, in this article we argue that, in the case of problems in which the data contain unrepresentative features, such as in EEG classification problems, it is beneficial to choose a much larger number of hidden nodes. We characterize this phenomenon, explain why this happens and exhibit several concrete examples to illustrate how ELMs behave. Furthermore, as searching for the optimal number of hidden nodes could be time consuming in enlarged ELMs, we propose a new training scheme, including a novel pruning method. This scheme provides an efficient way of finding the optimal number of nodes, making ELMs more suitable for dealing with real time EEG classification problems. Experimental results using synthetic data and real EEG data show a major improvement in the training time with respect to most traditional and state of the art ELM approaches, without jeopardising classification performance and resulting in more compact networks.

Assessment of Machine Learning Techniques for Oil Rig Classification in C-Band SAR Images

This article aims at performing maritime target classification in SAR images using machine learning (ML) and deep learning (DL) techniques. In particular, the targets of interest are oil platforms and ships located in the Campos Basin, Brazil. Two convolutional neural networks (CNNs), VGG-16 and VGG-19, were used for attribute extraction. The logistic regression (LR), random forest (RF), support vector machine (SVM), k-nearest neighbours (kNN), decision tree (DT), naive Bayes (NB), neural networks (NET), and AdaBoost (ADBST) schemes were considered for classification. The target classification methods were evaluated using polarimetric images obtained from the C-band synthetic aperture radar (SAR) system Sentinel-1. Classifiers are assessed by the accuracy indicator. The LR, SVM, NET, and stacking results indicate better performance, with accuracy ranging from 84.1% to 85.5%. The Kruskal–Wallis test shows a significant difference with the tested classifier, indicating that some classifiers present different accuracy results. The optimizations provide results with more significant accuracy gains, making them competitive with those shown in the literature. There is no exact combination of methods for SAR image classification that will always guarantee the best accuracy. The optimizations performed in this article were for the specific data set of the Campos Basin, and results may change depending on the data set format and the number of images.

Intelligent breast cancer diagnostic system empowered by deep extreme gradient descent optimization

Cancer is a manifestation of disorders caused by the changes in the body's cells that go far beyond healthy development as well as stabilization. Breast cancer is a common disease. According to the stats given by the World Health Organization (WHO), 7.8 million women are diagnosed with breast cancer. Breast cancer is the name of the malignant tumor which is normally developed by the cells in the breast. Machine learning (ML) approaches, on the other hand, provide a variety of probabilistic and statistical ways for intelligent systems to learn from prior experiences to recognize patterns in a dataset that can be used, in the future, for decision making. This endeavor aims to build a deep learning-based model for the prediction of breast cancer with a better accuracy. A novel deep extreme gradient descent optimization (DEGDO) has been developed for the breast cancer detection. The proposed model consists of two stages of training and validation. The training phase, in turn, consists of three major layers data acquisition layer, preprocessing layer, and application layer. The data acquisition layer takes the data and passes it to preprocessing layer. In the preprocessing layer, noise and missing values are converted to the normalized which is then fed to the application layer. In application layer, the model is trained with a deep extreme gradient descent optimization technique. The trained model is stored on the server. In the validation phase, it is imported to process the actual data to diagnose. This study has used Wisconsin Breast Cancer Diagnostic dataset to train and test the model. The results obtained by the proposed model outperform many other approaches by attaining 98.73 % accuracy, 99.60% specificity, 99.43% sensitivity, and 99.48% precision.

A Novel Key Features Screening Method Based on Extreme Learning Machine for Alzheimer's Disease Study

The Extreme Learning Machine (ELM) is a simple and efficient Single Hidden Layer Feedforward Neural Network(SLFN) algorithm. In recent years, it has been gradually used in the study of Alzheimer's disease (AD). When using ELM to diagnose AD based on high-dimensional features, there are often some features that have no positive impact on the diagnosis, while others have a significant impact on the diagnosis. In this paper, a novel Key Features Screening Method based on Extreme Learning Machine (KFS-ELM) is proposed. It can screen for key features that are relevant to the classification (diagnosis). It can also assign weights to key features based on their importance. We designed an experiment to screen for key features of AD. A total of 920 key functional connections screened from 4005 functional connections. Their weights were also obtained. The results of the experiment showed that: (1) Using all (4,005) features to diagnose AD, the accuracy is 95.33%. Using 920 key features to diagnose AD, the accuracy is 99.20%. The 3,085 (4,005 - 920) features that were screened out had a negative effect on the diagnosis of AD. This indicates the KFS-ELM is effective in screening key features. (2) The higher the weight of the key features and the smaller their number, the greater their impact on AD diagnosis. This indicates that the KFS-ELM is rational in assigning weights to the key features for their importance. Therefore, KFS-ELM can be used as a tool for studying features and also for improving classification accuracy.