A novel double deep ELMs ensemble system for time series forecasting

Boosting ridge for the extreme learning machine globally optimised for classification and regression problems

This paper explores the boosting ridge (BR) framework in the extreme learning machine (ELM) community and presents a novel model that trains the base learners as a global ensemble. In the context of Extreme Learning Machine single-hidden-layer networks, the nodes in the hidden layer are preconfigured before training, and the optimisation is performed on the weights in the output layer. The previous implementation of the BR ensemble with ELM (BRELM) as base learners fix the nodes in the hidden layer for all the ELMs. The ensemble learning method generates different output layer coefficients by reducing the residual error of the ensemble sequentially as more base learners are added to the ensemble. As in other ensemble methodologies, base learners are selected until fulfilling ensemble criteria such as size or performance. This paper proposes a global learning method in the BR framework, where base learners are not added step by step, but all are calculated in a single step looking for ensemble performance. This method considers (i) the configurations of the hidden layer are different for each base learner, (ii) the base learners are optimised all at once, not sequentially, thus avoiding saturation, and (iii) the ensemble methodology does not have the disadvantage of working with strong classifiers. Various regression and classification benchmark datasets have been selected to compare this method with the original BRELM implementation and other state-of-the-art algorithms. Particularly, 71 datasets for classification and 52 for regression, have been considered using different metrics and analysing different characteristics of the datasets, such as the size, the number of classes or the imbalanced nature of them. Statistical tests indicate the superiority of the proposed method in both regression and classification problems in all experimental scenarios.

A Hybrid System Based on Dynamic Selection for Time Series Forecasting

Hybrid systems, which combine statistical and machine learning (ML) techniques using residual (error forecasting) modeling, have been highlighted in the literature due to their accuracy and ability to forecast time series with different characteristics. In these architectures, a crucial task is the proper modeling of the residuals since they may present random fluctuations, complex nonlinear patterns, and heteroscedastic behavior. Hence, the selection, specification, and training of one ML model to forecast the residuals are costly and challenging tasks since issues, such as underfitting, overfitting, and misspecification, can lead to a system with low accuracy or even deteriorate the linear forecast of the time series. This article proposes a hybrid system, named dynamic residual forecasting (DReF), that employs a modified dynamic selection (DS) algorithm to decide: the most suitable ML model to forecast a pattern of the residual series and if it is a promising candidate to increase the accuracy of the time series forecast from the linear combination. Thus, the DReF aims to reduce the uncertainty of the ML model selection and avoid the deterioration of the time series forecast. Furthermore, the proposed system searches for the most suitable parameters of the DS algorithm for each data set. In this article, the proposed method uses a pool of five ML models widely adopted in the literature: multilayer perceptron, support vector regression, radial basis function, long short-term memory, and convolutional neural network. An experimental evaluation was conducted using ten well-known time series. The results show that the DReF obtains superior results for the majority of the data sets compared with single and hybrid models of the literature.

A Fetal ECG Extraction Method Based on ELM Optimized by Improved PSO Algorithm

The extraction of fetal electrocardiogram (FECG) is of great significance for perinatal fetal monitoring. In order to improve the prediction accuracy of FECG, a FECG extraction method based on extreme learning machine (ELM) optimized by an improved particle swarm optimization (IPSO) was proposed (IPSO-ELM). First, according to the characteristics of the mixed signal on the maternal abdominal wall, and based on the global search ability of IPSO, the initial weight matrix and hidden layer bias of ELM were optimized to match with the mixed signal of the maternal abdominal wall and the network topology. Then, an ELM model was established using the optimal network parameters obtained by IPSO. The nonlinear transformation of the maternal ECG (MECG) signal to the abdominal wall was estimated by the IPSO-ELM network. Finally, the non-linearly transformed MECG signal was mixed with the abdominal wall subtract to obtain clear FECG. The experimental results of clinical ECG signals in DaISy dataset showed that, compared with the traditional normalized minimum mean square error, support vector machine method, and ELM neural network methods, the proposed method can extract clearer FECG signals and improve the signal-to-noise ratio of extracted FECG.

Imbalanced biomedical data classification using self-adaptive multilayer ELM combined with dynamic GAN

BACKGROUND

Imbalanced data classification is an inevitable problem in medical intelligent diagnosis. Most of real-world biomedical datasets are usually along with limited samples and high-dimensional feature. This seriously affects the classification performance of the model and causes erroneous guidance for the diagnosis of diseases. Exploring an effective classification method for imbalanced and limited biomedical dataset is a challenging task.

METHODS

In this paper, we propose a novel multilayer extreme learning machine (ELM) classification model combined with dynamic generative adversarial net (GAN) to tackle limited and imbalanced biomedical data. Firstly, principal component analysis is utilized to remove irrelevant and redundant features. Meanwhile, more meaningful pathological features are extracted. After that, dynamic GAN is designed to generate the realistic-looking minority class samples, thereby balancing the class distribution and avoiding overfitting effectively. Finally, a self-adaptive multilayer ELM is proposed to classify the balanced dataset. The analytic expression for the numbers of hidden layer and node is determined by quantitatively establishing the relationship between the change of imbalance ratio and the hyper-parameters of the model. Reducing interactive parameters adjustment makes the classification model more robust.

RESULTS

To evaluate the classification performance of the proposed method, numerical experiments are conducted on four real-world biomedical datasets. The proposed method can generate authentic minority class samples and self-adaptively select the optimal parameters of learning model. By comparing with W-ELM, SMOTE-ELM, and H-ELM methods, the quantitative experimental results demonstrate that our method can achieve better classification performance and higher computational efficiency in terms of ROC, AUC, G-mean, and F-measure metrics.

CONCLUSIONS

Our study provides an effective solution for imbalanced biomedical data classification under the condition of limited samples and high-dimensional feature. The proposed method could offer a theoretical basis for computer-aided diagnosis. It has the potential to be applied in biomedical clinical practice.

Research on air pollutant concentration prediction method based on self-adaptive neuro-fuzzy weighted extreme learning machine

In order to improve the prediction accuracy and real-time of the air pollutant concentration prediction, this paper proposes self-adaptive neuro-fuzzy weighted extreme learning machine (ANFIS-WELM) based on the weighted extreme learning machine (WELM) and the adaptive neuro-fuzzy inference system (ANFIS) combined air pollutant concentration prediction method. Firstly, Gaussian membership function parameters are selected to fuzzify the input values and calculate the membership degree of each input variable. Secondly, corresponding fuzzy rules are activated, and the firing strength is normalized to calculate the output matrix of hidden nodes. Then, the optimal parameters (C, M), weights are assigned to weighted ELM by using locally weighted linear regression, and the regularized WELM target formula with equality constraint is optimized by the Karush-Kuhn-Tucker (KKT) conditions, the output weight matrix is calculated, and finally the prediction output matrix is calculated. Based on the air pollutant concentration data collected in Datong, Taiwan, the data on the pollutants containing carbon monoxide (CO), nitric oxide (NO), PM_2.5 (particulate matter) and PM₁₀, are selected by different historical time series lengths, using genetic algorithm-backpropagation neural network (GA-BPNN), support vector regression (SVR), extreme learning machine (ELM), WELM, ANFIS, regularized extreme learning adaptive neuro-fuzzy inference system (R-ELANFIS) and ANFIS-WELM are built for predict the concentration of each pollutant collected by single monitoring point in single-step time series. The experimental results show that the ANFIS-WELM presented in this paper has better prediction accuracy and real-time performance, realizes the prediction of multi-step time series on the basis of the ANFIS-WELM, and realizes the engineering application of the ANFIS-WELM algorithm package on the self-developed mobile source emissions online monitoring data center software system.

Research and application of a novel hybrid air quality early-warning system: A case study in China

As one of the most serious meteorological disasters in modern society, air pollution has received extensive attention from both citizens and decision-makers. With the complexity of pollution components and the uncertainty of prediction, it is both critical and challenging to construct an effective and practical early-warning system. In this paper, a novel hybrid air quality early-warning system for pollution contaminant monitoring and analysis was proposed. To improve the efficiency of the system, an advanced attribute selection method based on fuzzy evaluation and rough set theory was developed to select the main pollution contaminants for cities. Moreover, a hybrid model composed of the theory of "decomposition and ensemble", an extreme learning machine and an advanced heuristic algorithm was developed for pollution contaminant prediction; it provides deterministic and interval forecasting for tackling the uncertainty of future air quality. Daily pollution contaminants of six major cities in China were selected as a dataset to evaluate the practicality and effectiveness of the developed air quality early-warning system. The superior experimental performance determined by the values of several error indexes illustrated that the proposed early-warning system was of great effectiveness and efficiency.