Random Vector Functional Link Research Articles

Intelligent Feature Selection with Deep Learning Based Financial Risk Assessment Model

Due to global financial crisis, risk management has received significant attention to avoid loss and maximize profit in any business. Since the financial crisis prediction (FCP) process is mainly based on data driven decision making and intelligent models, artificial intelligence (AI) and machine learning (ML) models are widely utilized. This article introduces an intelligent feature selection with deep learning based financial risk assessment model (IFSDL-FRA). The proposed IFSDL-FRA technique aims to determine the financial crisis of a company or enterprise. In addition, the IFSDL-FRA technique involves the design of new water strider optimization algorithm based feature selection (WSOA-FS) manner to an optimum selection of feature subsets. Moreover, Deep Random Vector Functional Link network (DRVFLN) classification technique was applied to properly allot the class labels to the financial data. Furthermore, improved fruit fly optimization algorithm (IFFOA) based hyperparameter tuning process is carried out to optimally tune the hyperparameters of the DRVFLN model. For enhancing the better performance of the IFSDL-FRA technique, an extensive set of simulations are implemented on benchmark financial datasets and the obtained outcomes determine the betterment of IFSDL-FRA technique on the recent state of art approaches.

Application of nonlinear system identification for EEG modelling using VMD-based deep random vector functional link network

Application of nonlinear system identification for EEG modelling using VMD-based deep random vector functional link network

Bayesian neural network with efficient priors for online quality prediction

In process industries, neural networks are popular models for quality prediction due to their superiorities on dealing with nonlinearity and high dimensionality of process data. However, small sample size of online data will result in epistemic uncertainty of prediction model, so neural networks with fixed parameters are prone to overfitting and provide over-confident predictions with poor robustness. To this end, we propose a Pior-Assistant Random Bayesian Neural Network (PA-RBNN). First, the Random Vector Functional Link Neural Network (RVFLNN) with Bayes backpropagation (RBNN) is introduced to the last layer of the off-lined deep neural network model. The prior mean values and variances of parameters are determined by the closed-form solution from RVFLNN. Then, the semi-supervised form of PA-RBNN is further developed by introducing manifold regularization to RVFLNN to obtain the semi-supervised prior means and variances of parameters. In this way, the proposed prior provides a reasonable range before learning the parameter posterior, and both the labeled and unlabeled data are utilized to obtain the prior. The proposed method is verified by the benchmarked industrial process data. The experimental results show that the assistant prior has an optimistic effect on guiding the training process towards the basins of attraction of minima with less computation cost, and improves the generalization of prediction model.

Randomized neural networks for multilabel classification

Multilabel classification is a supervised learning problem in which input instances belong to multiple output labels. In this paper, we propose noniterative randomization-based neural networks for multilabel classification. These multilabel neural networks are named as Multilabel Random Vector Functional Link Network (ML-RVFL), Multilabel Kernelized Random Vector Functional Link Network (ML-KRVFL), Multilabel Broad Learning System (ML-BLS), and Multilabel Fuzzy Broad Learning System (ML-FBLS). The output weights of these neural networks are computed using pseudoinverse. At the output layer, multilabel classification is performed by using an adaptive threshold function. The computation of output weights using pseudoinverse retains the faster computation power of these algorithms compared to iterative learning algorithms. The adaptive threshold function used in the proposed approach can consider the correlation among the output labels and the whole dataset for threshold computation. Five multilabel evaluation metrics evaluate the proposed multilabel neural networks on 12 benchmark datasets of various domains such as text, image, and genomics. The ML-KRVFL provides the overall best Friedman rankings on five evaluation metrics followed by ML-RVFL, ML-FBLS, and ML-BLS, respectively. Based on the experimentation results, the proposed ML-KRVFL, ML-RVFL, ML-FBLS, and ML-BLS perform better than other relevant multilabel approaches in the mentioned order.The proposed approaches are faster than other state-of-the-art iterative approaches and noniterative approaches in terms of running time.

Improving Drought Modeling Using Hybrid Random Vector Functional Link Methods

Drought modeling is essential in water resources planning and management in mitigating its effects, especially in arid regions. Climate change highly influences the frequency and intensity of droughts. In this study, new hybrid methods, the random vector functional link (RVFL) integrated with particle swarm optimization (PSO), the genetic algorithm (GA), the grey wolf optimization (GWO), the social spider optimization (SSO), the salp swarm algorithm (SSA) and the hunger games search algorithm (HGS) were used to forecast droughts based on the standard precipitation index (SPI). Monthly precipitation data from three stations in Bangladesh were used in the applications. The accuracy of the methods was compared by forecasting four SPI indices, SPI3, SPI6, SPI9, and SPI12, using the root mean square errors (RMSE), the mean absolute error (MAE), the Nash–Sutcliffe efficiency (NSE), and the determination coefficient (R2). The HGS algorithm provided a better performance than the alternative algorithms, and it considerably improved the accuracy of the RVFL method in drought forecasting; the improvement in RMSE for the SPI3, SP6, SPI9, and SPI12 was by 6.14%, 11.89%, 14.14%, 24.5% in station 1, by 6.02%, 17.42%, 13.49%, 24.86% in station 2 and by 7.55%, 26.45%, 15.27%, 13.21% in station 3, respectively. The outcomes of the study recommend the use of a HGS-based RVFL in drought modeling.

Predicting the mechanical properties of Cu–Al2O3 nanocomposites using machine learning and finite element simulation of indentation experiments

Micromechanics model, finite element (FE) simulation of microindentation and machine learning were deployed to predict the mechanical properties of Cu–Al2O3 nanocomposites. The micromechanical model was developed based on the rule of mixture and grain and grain boundary sizes evolution to predict the elastic modulus of the produced nanocomposites. Then, a FE model was developed to simulate the microindentation test. The input for the FE model was the elastic modulus that was computed using the micromechanics model and wide range of yield and tangent stresses values. Finally, the output load-displacement response from the FE model, the elastic modulus, the yield and tangent strengths used for the FE simulations, and the residual indentation depth were used to train the machine learning model (Random vector functional link network) for the prediction of the yield and tangent stresses of the produced nanocomposites. Cu–Al2O3 nanocomposites with different Al2O3 concentration were manufactured using insitu chemical method to validate the proposed model. After training the model, the microindentation experimental load-displacement curve for Cu–Al2O3 nanocomposites was fed to the machine learning model and the mechanical properties were obtained. The obtained mechanical properties were in very good agreement with the experimental ones achieving 0.99 coefficient of determination R2 for the yield strength.

Classification of clothing images based on a parallel convolutional neural network and random vector functional link optimized by the grasshopper optimization algorithm

To improve accuracy in clothing image recognition, this paper proposes a clothing classification method based on a parallel convolutional neural network (PCNN) combined with an optimized random vector functional link (RVFL). The method uses the PCNN model to extract features of clothing images. Then, the structure-intensive and dual-channel convolutional neural network (i.e., the PCNN) is used to solve the problems of traditional convolutional neural networks (e.g., limited data and prone to overfitting). Each convolutional layer is followed by a batch normalization layer, and the leaky rectified linear unit activation function and max-pooling layers are used to improve the performance of the feature extraction. Then, dropout layers and fully connected layers are used to reduce the amount of calculation. The last layer uses the RVFL as optimized by the grasshopper optimization algorithm to replace the SoftMax layer and classify the features, further improving the stability and accuracy of classification. In this study, two aspects of the classification (feature extraction and feature classification) are improved, effectively improving the accuracy. The experimental results show that on the Fashion-Mnist dataset, the accuracy of the algorithm in this study reaches 92.93%. This value is 1.36%, 2.05%, 0.65%, and 3.76% higher than that of the local binary pattern (LBP)-support vector machine (SVM), histogram of oriented gradients (HOG)-SVM, LBP-HOG-SVM, and AlexNet-sparse representation-based classifier algorithms, respectively, effectively demonstrating the classification performance of the algorithm.

Android malware classification based on random vector functional link and artificial Jellyfish Search optimizer.

Smartphone usage is nearly ubiquitous worldwide, and Android provides the leading open-source operating system, retaining the most significant market share and active user population of all open-source operating systems. Hence, malicious actors target the Android operating system to capitalize on this consumer reliance and vulnerabilities present in the system. Hackers often use confidential user data to exploit users for advertising, extortion, and theft. Notably, most Android malware detection tools depend on conventional machine-learning algorithms; hence, they lose the benefits of metaheuristic optimization. Here, we introduce a novel detection system based on optimizing the random vector functional link (RVFL) using the artificial Jellyfish Search (JS) optimizer following dimensional reduction of Android application features. JS is used to determine the optimal configurations of RVFL to improve classification performance. RVFL+JS minimizes the runtime of the execution of the optimized models with the best performance metrics, based on a dataset consisting of 11,598 multi-class applications and 471 static and dynamic features.

Illumination estimation via random vector functional link based on improved arithmetic optimization algorithm

AbstractIn order to solve the problem that the existing illumination estimation methods have low accuracy in image restoration, a random vector functional link illumination estimation algorithm based on whale optimization arithmetic algorithm is proposed in this article. First, in order to improve the optimization accuracy of arithmetic optimization algorithm, the whale optimization algorithm is applied to optimize the initial population of the arithmetic optimization algorithm. Then, we utilize the improved arithmetic optimization algorithm to improve the input weight and hidden layer offset of random vector functional link, and exclude the uncertainty caused by the random generation of random vector functional link's input weight and hidden layer bias. Finally, the proposed whale arithmetic optimization algorithm‐random vector functional link model is applied to SFU lab dataset and Gehler‐Shi dataset for illumination estimation. Compared with the other earlier illumination estimation algorithms, the proposed illumination estimation model in this article has better effect and higher stability. In addition, due to the diagonal mapping method for color correction and image restoration, the proposed illumination estimation algorithm can also get better results in image restoration.

High-accuracy gearbox health state recognition based on graph sparse random vector functional link network

Health state recognition technology is of great significance for the maintenance decision-making and safety assessment of gearboxes. Traditional random vector functional link network (RVFLN) cannot fully leverage the label sparsity of gearbox health states and the manifold structure information of raw data is neglected. To overcome these drawbacks, a high-accuracy gearbox health state recognition model is proposed in this paper, namely graph sparse RVFLN (GSRVFLN). Firstly, a sparse constraint term is introduced into GSRVFLN to force the predicted label to be similar to the zero-one true label as much as possible, so as to fully exploit the sparsity of the output label. Secondly, a discriminative adjacency graph is designed for GSRVFLN with the label information to capture the inherent geometry structure and discriminative information of data. Finally, we derive an effective solution for GSRVFLN with the alternating direction method of multipliers (ADMM) framework, and this solution has great convergence. The applicability of GSRVFLN for health state recognition is validated with two gearbox datasets, and comparative results show that GSRVFLN achieves the excellent performance of gearbox health state recognition, winning other state-of-the-art models.

Identification of wrist movements based on magnetoencephalograms via noise assisted multivariate empirical mode decomposition

Brain computer interface (BCI) is a medium that converts the brain activity signals such as the electroencephalograms (EEGs) to the motion control signals such as the wrist movement signals. Recently, the magnetoencephalograms (MEGs) are used to record the brain activities representing the wrist movements of the healthy right handed subjects. Here, there are four types of the wrist movements. They are the right movements, the forward movements, the left movements and the back movements of the wrist. Since the MEGs are not the monotonic frequency signals, one of the major challenges is the difficulty to extract the features from the pieces of a finite duration of the MEGs for classifying the wrist movements.In order to overcome this challenge, the noiseassisted multivariate empirical mode decomposition (NA-MEMD) is proposed. There are three major steps for our proposed NA-MEMD based algorithm. The first step is to employ the NA-MEMD for performing the multi-channel and the multi-scale signal denoising. The second step isto extract the statistical features such as the mean and the variance as well as the time frequency features such as the marginal spectrum entropy, the multi-scale permutation entropy and the multi-scale fuzzy entropy. Finally, some typical classifiers such as the random forest (RF),the back propagation neural network (BPNN) and the support vector machine (SVM) as well as some advanced classifiers such as the extreme learning machine (ELM), the random vector functional link (RVFL) and the long short term memory network (LSTM) are employed for performing the classification. The computer numerical simulation results show that our proposed NA-MEMD based algorithm achieves a higher classification accuracy compared to the state of the artmethods without the multi-channel and the multi-scale analysis and the semi-improvedmethods with either the multi-channel or the multi-scale analysis.

Reinforced fuzzy clustering-based rule model constructed with the aid of exponentially weighted ℓ2 regularization strategy and augmented random vector functional link network

Fuzzy rule-based models are widely employed to tackle regression problems due to their simplicity and comprehensibility. Numeric functions (e.g., linear ones) are generally utilized to represent the conclusion part of a rule, but this simple module cannot well describe the behavior of data located within local area defined by the premise part. In this study, a novel reinforced fuzzy clustering-based rule model (RFCRM) is presented to address this issue. First, information granulation (Fuzzy C-Means) approach is used to reveal the structure in the data and divide the input space into local regions, as well as form the premise parts. Second, the augmented Random Vector Functional Link Networks (RVFLNs) are employed as the conclusion parts to strengthen the description and representation of the behavior of data positioned within the local regions. Weighted Least Squares Error estimation with regularization is exploited to estimate the coefficients of the connection weights of RVFLN. Also, an exponential weight method benefited from the weight function theory encountered in harmonic analysis is introduced to improve the penalty strategy of regularization in coefficient estimation. More specifically, an exponential weighted ℓ2 regularization (EWℓ2) strategy equipped with the exponential functions as the penalty terms is proposed to realize the targeted penalty on coefficients. The merit of EWℓ2 over ordinary ℓ2 is that different types of polynomial terms can be identified and penalized separately in coefficient estimation. This strategy not only prevents the decline of generalization ability, but also effectively enhances the prediction potential of the model. By combining weighted LSE based on EWℓ2 with FCM partition and the augmented RVFLN leads to better prediction accuracy as well as lower complexity (viz., number of rules). Finally, detailed experimental studies are conducted to demonstrate the effectiveness of the proposed RFCRM.

PBTNet: A New Computer-Aided Diagnosis System for Detecting Primary Brain Tumors.

Brain tumors are among the leading human killers. There are over 120 different types of brain tumors, but they mainly fall into two groups: primary brain tumors and metastatic brain tumors. Primary brain tumors develop from normal brain cells. Early and accurate detection of primary brain tumors is vital for the treatment of this disease. Magnetic resonance imaging is the most common method to diagnose brain diseases, but the manual interpretation of the images suffers from high inter-observer variance. In this paper, we presented a new computer-aided diagnosis system named PBTNet for detecting primary brain tumors in magnetic resonance images. A pre-trained ResNet-18 was selected as the backbone model in our PBTNet, but it was fine-tuned only for feature extraction. Then, three randomized neural networks, Schmidt neural network, random vector functional-link, and extreme learning machine served as the classifiers in the PBTNet, which were trained with the features and their labels. The final predictions of the PBTNet were generated by the ensemble of the outputs from the three classifiers. 5-fold cross-validation was employed to evaluate the classification performance of the PBTNet, and experimental results demonstrated that the proposed PBTNet was an effective tool for the diagnosis of primary brain tumors.

Illumination correction with optimized kernel extreme learning machine based on improved marine predators algorithm

AbstractAiming at the problem of poor image illumination correction accuracy, a kernel extreme learning machine optimized based on the differential evolution‐improved marine predators algorithm is proposed to estimate the scene illumination information and restore the image. First, in order to solve the problem of randomization of the initial population of the marine predators algorithm, differential evolutions is used to provide a set of suitable initial populations to the algorithm. Then, an improved algorithm is used to optimize the weight and bias of the kernel extreme learning machine to solve the problem of the randomness of the weight and bias, so that the learning machine converges to the global optimal value and avoids the unstable predictions. Finally, after obtaining the predicted illumination information, the image model is corrected to the image effect under standard illumination through diagonal transformation. It can be seen from the experimental results that the best predictive value of the differential evolution marine predator algorithm learning machine is 0.0135915, and the quasi‐bias is only 0.001687. Compared with traditional illumination estimation algorithms such as random vector functional link and extreme learning machine, the differential evolution marine predator algorithm learning machine algorithm proposed in this article has better stability, higher accuracy of calculated predicted values, and better image restoration effect.

A digital direction of arrival estimator based on fast on-line sequential random vector functional link network

This article develops a fast online sequential random vector functional network. In this network, input nodes are connected to both hidden nodes and output nodes. This architecture results in high-speed learning and testing. The network estimates the direction of arrival from signals received by the sparse array in real-time. Neither noisy signal nor effect of mutual coupling among array elements affect the estimation appreciably using the network. The network shows competitive performance in terms of learning speed, accuracy, short event regression time, simplicity, and robustness in comparison to other state-of-the-art methods such as support vector regression, extreme learning machine, and online sequential extreme learning machine. Finally, validation of the network translation into a hardware platform using a fast digital high-speed reconfigurable Xilinx Virtex-5 field-programmable gate array embedded processor for single-source detection completes the work. Comparison with reported measurement shows results with good accuracy and low root mean square error both by the network and its hardware-implemented platform. The direction of arrival estimation is thus possible using the network either through a computer or dedicated hardware as per necessity.

Application of Machine Learning Algorithms for Geogenic Radon Potential Mapping in Danyang-Gun, South Korea

Continuous generation of radon gas by soil and rocks rich in components of the uranium chain, along with prolonged inhalation of radon progeny in enclosed spaces, can lead to severe respiratory diseases. Detection of radon-prone areas and acquisition of detailed knowledge regarding relationships between indoor radon variations and geogenic factors can facilitate the implementation of more appropriate radon mitigation strategies in high-risk residential zones. In the present study, 10 factors (i.e., lithology; fault density; mean soil calcium oxide [CaO], copper [Cu], lead [Pb], and ferric oxide [Fe2O3] concentrations; elevation; slope; valley depth; and the topographic wetness index [TWI]) were selected to map radon potential areas based on measurements of indoor radon levels in 1,452 dwellings. Mapping was performed using three machine learning methods: long short-term memory (LSTM), extreme learning machine (ELM), and random vector functional link (RVFL). The results were validated in terms of the area under the receiver operating characteristic curve (AUROC), root mean square error (RMSE), and standard deviation (StD). The prediction abilities of all models were satisfactory; however, the ELM model had the best performance, with AUROC, RMSE, and StD values of 0.824, 0.209, and 0.207, respectively. Moreover, approximately 40% of the study area was covered by very high and high-risk radon potential zones that mainly included populated areas in Danyang-gun, South Korea. Therefore, the map can be used to establish more appropriate construction regulations in radon-priority areas, and identify more cost-effective remedial actions for existing buildings, thus reducing indoor radon levels and, by extension, radon exposure-associated effects on human health.

Hierarchical polynomial-based fuzzy neural networks driven with the aid of hybrid network architecture and ranking-based neuron selection strategies

In this study, we propose hierarchical polynomial-based fuzzy neural networks (HPFNN). The aim of this study is to develop the design methodologies of hierarchical model to improve the prediction accuracy of the model without sacrificing computational efficiency through combining hybrid network architecture composed of different neurons and ranking-based neuron selection strategies. The essential bullets of the proposed model can be enumerated as follows: (a) A hybrid network architecture is designed by combining the traits of fuzzy rule-based neurons with random vector functional link (FRN-RVFL) and polynomial neurons. FRN-RVFL is employed to erect the first layer of network. The entire network topology and the neurons of the remaining layers are constructed with polynomial neural network. (b) Two kinds of ranking-based neuron selection (RNS) strategies such as Linear-RNS (LRNS) and Exponential-RNS (EPNS) are presented. Compared with traditional neuron selection strategy, RNS can enrich the diversity of candidate neurons while maintaining the approximation ability of neurons, which provides opportunities for selecting neurons with predictive potential. (c) Regularization-based least square approach is applied to alleviate the possible overfitting in coefficient estimation as well as enhance generalization abilities of the model. The performance of HPFNN is verified by using a series of synthetic data and machine learning datasets. Based on the experimental results, HPFNN exhibits sound prediction accuracy and reasonable computational cost in contrast with the same type of hierarchical models. Furthermore, HPFNN achieves better generalization performance on at least 12 of 20 datasets when compared with the performance of state-of-the-art models.

Modeling of drilling process of GFRP composite using a hybrid random vector functional link network/parasitism-predation algorithm

Non-laminated glass fiber-reinforced epoxy composites (GFREC) have shown promising applications in various engineering fields. In this study, an experimental investigation followed by artificial intelligent modeling is carried out on the drilling of GFREC under two cooling conditions, namely internal and external. The damage factor of the drilled holes and the temperature of the drill tip were considered as main process responses. The experiments were conducted under different combinations of feed, spindle speed and coolant pressure. All holes were drilled using fresh tungsten carbide twist drills coated by TiN/TiAlN layer. A fine-tuned random vector functional link network (RVFL) model incorporated with a new optimizer called parasitism-predation algorithm (PPA) was developed to model the drilling process. PPA acts as an advanced metaheuristic optimizer to obtain the best model variables of RVFL. The developed model was compared with the stand-alone RVFL as well as the fined-tuned RVFL using particle swarm optimizer (PSO). RVFL-PPA outperformed other models in terms of coefficient of determination, root mean square error and other measures. The coefficient of determination of RVFL-PPA for different responses and cooling conditions ranges between 0.995 and 1.

A new fine-tuned random vector functional link model using Hunger games search optimizer for modeling friction stir welding process of polymeric materials

Modeling of manufacturing processes using artificial intelligence-based techniques has recently received considerable attention. The current investigation introduces a new artificial intelligence-based predictive model for friction stir welding of dissimilar polymeric materials. The welded joint is made of acrylonitrile butadiene styrene (ABS) and polycarbonate (PC) sheets. The proposed model is used to correlate the joint characteristics (tensile strength, joint efficiency and extensibility) with the welding variables (rotational tool speed, welding speed and tool tilt angle). The model consists of a random vector functional link (RVFL) model optimized by Hunger games search (HGS) optimizer. HGS optimizer is utilized to find out the optimal internal parameters of RVFL that boost the model accuracy. The proposed model is compared with two other optimized models in which RVFL is integrated with the sine cosine algorithm (SCA) or ecosystem-based optimizer (AEO). RVFL-HGS outperformed RVFL-SCA and RVFL-AEO based on different statistical measures. For all investigated joint characteristics, the determination coefficient ranges between 0.907 and 0.993 for RVFL-HGS, 0.869 and 0.972 for RVFL-AEO, 0.829 and 0.983 for RVFL-SCA.

Time series classification using diversified Ensemble Deep Random Vector Functional Link and Resnet features

Random Vector Functional Link (RVFL) is popular among researchers in many areas of machine learning. RVFL is preferred by many researchers as RVFL can produce good performance with relatively little training time. Recent works extend RVFL into deep and ensemble versions. However, RVFL does not have effective feature extraction methods commonly used in time series classification. This results in poor performance of RVFL in time series classification tasks. Also, deep RVFL is a relatively new and evolving area of research. In this paper, we present a framework that extracts features from Residual Networks (Resnet) and trains Ensemble Deep Random Vector Functional Link (edRVFL). We use features extracted from every residual block to train an ensemble of edRVFLs. We propose the following enhancements to edRVFL. Firstly, we diversity the structure of edRVFL and the direct link features to encourage diversity. Secondly, we built an ensemble of edRVFLs with the top two activation functions. Thirdly, we use two-stage tuning to save computational costs. Lastly, we perform a weighted average of all decisions made by every edRVFL. Experiments on the 55 largest UCR datasets show that using features extracted from all Residual blocks improves performance. All our proposed enhancements help improve classification accuracy or computational effort. Consequently, our proposed framework outperforms all traditional and deep learning-based time series classification methods.