Different Number Of Hidden Nodes Research Articles

Diabetes detection based on Fast Learning Network model

In the last decades, machine learning algorithms have witnessed a high significance in healthcare applications in terms of detecting various diseases. For instance, diabetic disease is considered one of the major health problems around the world. However, there is a need to deeply study a machine learning algorithm in the detection of diabetic disease. Therefore, this study presents a Fast Learning Network (FLN) algorithm in the detection of diabetic disease based on different numbers of hidden nodes. In this work, the Pima Indians Diabetes Database (PIDD) is used for training and testing the proposed FLN algorithm. Furthermore, the performance of the proposed model has been assessed in terms of several evaluation measurements such as accuracy, precision, recall, F-Measure, G-Mean, MCC, and specificity. The experimental results show that the highest achieved accuracy, recall, F-Measure, G-Mean, and MCC were 82.17%, 80.95%, 71.33%, 71.84%, and 59.54%, respectively. Meanwhile, the highest obtained results for precision and specificity were 67.50% and 83.12%, respectively. In addition, the performance of the proposed model has outperformed its comparative in terms of detection accuracy.

Spatial Prediction of Monthly Precipitation in Sulaimani Governorate using Artificial Neural Network Models

ANN modeling is used here to predict missing monthly precipitation data in one station of the eight weather stations network in Sulaimani Governorate. Eight models were developed, one for each station as for prediction. The accuracy of prediction obtain is excellent with correlation coefficients between the predicted and the measured values of monthly precipitation ranged from (90% to 97.2%). The eight ANN models are found after many trials for each station and those with the highest correlation coefficient were selected. All the ANN models are found to have a hyperbolic tangent and identity activation functions for the hidden and output layers respectively, with learning rate of (0.4) and momentum term of (0.9), but with different data set sub-division into training, testing and holdout data sub-sets, and different number of hidden nodes in the hidden layer. It is found that it is not necessary that the nearest station to the station under prediction has the highest effect; this may be attributed to the high differences in elevation between the stations. It can also found that the variance is not necessary has effect on the correlation coefficient obtained.

Neural Network Model in Forecasting Malaysia’s Unemployment Rates

Neural networks (NN) have been widely applied in time series forecasting. This study aims to develop basic NN models for forecasting the unemployment rate in Malaysia by gender. The yearly unemployment rate of thirty-eight years from the year 1982 to 2019 was obtained from the Department of Statistics Malaysia. In addition, datasets of gross domestic product, inflation and population rates extracted from the World Bank Data website were used as input variables in developing the NN models. Several NN models with different number of hidden nodes were developed and evaluated. Results showed that the best model for the male population was the NN model with four hidden nodes in one hidden layer whereas the NN model with two hidden nodes in one hidden layer was the best for the female population. Additionally, it can be concluded that the trend for the future unemployment rate in Malaysia for male and female population in the next ten years will be gradually constant throughout the year starting from 2020 to 2030.

Predicting Dengue Outbreak based on Meteorological Data Using Artificial Neural Network and Decision Tree Models

Dengue fever is well-known as a potentially fatal disease, and the number of cases in some areas remains uncontrolled. Despite efforts to prevent the dengue outbreak from spreading further, vectors may be to blame. Identifying what weather characteristics contribute to dengue outbreaks is important to predict the dengue outbreak. This study proposes Artificial Neural Network (ANN) and Decision Tree (DT) models based on maximum temperature, minimum temperature, total rainfall, and average humidity to predict the dengue outbreak in Kota Bharu. Different numbers of hidden nodes were used in ANN to optimize the model. Both models, ANN and DT are evaluated based on accuracy, sensitivity and specificity showing that ANN (Accuracy = 68.85%, Sensitivity = 99.71%, Specificity = 1.27%), performed better than DT (Accuracy = 67.46%, Sensitivity = 98.82%, Specificity = 2.53%). This means that ANN outperforms DT when predicting a dengue outbreak in Kota Bharu. Based on the ANN model, it can be concluded that the number of hidden nodes affects the model's accuracy. Selecting the ideal number of hidden nodes for modeling the ANN model is appropriate. Even though ANN accuracy for prediction models is greater than DT, it is still low. It can be inferred that selecting a prediction model appropriate for a variety of dataset types and levels of complexity is important. Based on these models, the government may take pre-emptive actions to enhance public awareness about climate change.

Fusion Evaluation of College Cultivation by Adaptive Multivariate Neural Network Model.

The quality of graduates is the key factor in evaluating the cultivation effect of colleges and universities. Quantification of whether the graduates qualify for their working post in companies and industries provides conduction for further college cultivation reform enhancement. In this work, we proposed an adaptive multivariate neural network architecture for fusion evaluation of college student cultivation. Specifically, we designed a questionnaire to collect data on the current working status of 1231 graduates and recorded 32 in-school training items categorized into four different modules. For quantitative evaluation, 10 indices of career-require competence were set to describe the graduates' job abilities. The fused contribution of the in-school training items to the career-required competence was predicted by the multivariate network model with the linking weights adaptively trained. A comprehensive contribution matrix was generated by discrete PCA multivariate transforming to provide a digital reference for the network training. A 7-level scoring system was designed for quantifying the contribution matrix. For model optimization, the network structure was tuned by testing a different number of hidden nodes. The model was trained and optimized to reveal the direct correlation between college cultivation and job-required abilities. Experimental results indicated that the methodology we proposed is feasible to evaluate the cultivation mode in colleges and universities, theoretically and technically providing positive directions for colleges and universities to make their cultivation reforming, as to enhance the quality of their graduates.

Characters, Spirals and Unit Dynamics by training neural network model

Deep learning is an integral part of machine learning by building a structure like the neural network in the human brain. Amounts of different neural network models have been emerged, improved, and transformed. This article will implement various neural network models for three different specific tasks (Japanese character recognition/double helix task/remote unit dynamics) and analyse the fine-tuning parameters process (different network layers, different types of network layers). The result is that generally, models with more layers would perform better. However, other parameter could have significant effects on the performance. For example, activation function, different numbers of hidden nodes, different initial weights, and different periods, etc.

Bayesian robust multi-extreme learning machine

Outliers usually exist in the collected data due to human or instrumentation errors. Their existence makes the regression error obey a non-Gaussian distribution. So, people assume that the regression error obeys finite mixture distributions, such as mixture of Gaussians and mixture of Student’s t-distributions. This paper proposed two Bayesian robust regression models based on multiple extreme learning machines. First, the finite mixture error distributions are extended to their infinite counterparts according to the theory of stick-breaking construction, which can avoid selecting the optimal number of components in mixture models. Second, a multi-extreme learning machine, which combines many single extreme learning machines with different numbers of hidden nodes, is constructed, which can avoid the effects of different numbers of hidden nodes on the final regression performance. Besides, sparse Bayesian learning has also been derived to perform the sparse model weight and output weight inference automatically. The experimental results on artificial datasets and eight benchmark datasets show that the proposed two models outperform the other compared models under different rates of outliers. Also, they generate better multi-step ahead wind speed and traffic speed forecasts in real applications.

Quantitative thickness prediction of tectonically deformed coal using Extreme Learning Machine and Principal Component Analysis: a case study

The thickness of tectonically deformed coal (TDC) has positive correlation associations with gas outbursts. In order to predict the TDC thickness of coal beds, we propose a new quantitative predicting method using an extreme learning machine (ELM) algorithm, a principal component analysis (PCA) algorithm, and seismic attributes. At first, we build an ELM prediction model using the PCA attributes of a synthetic seismic section. The results suggest that the ELM model can produce a reliable and accurate prediction of the TDC thickness for synthetic data, preferring Sigmoid activation function and 20 hidden nodes. Then, we analyze the applicability of the ELM model on the thickness prediction of the TDC with real application data. Through the cross validation of near-well traces, the results suggest that the ELM model can produce a reliable and accurate prediction of the TDC. After that, we use 250 near-well traces from 10 wells to build an ELM predicting model and use the model to forecast the TDC thickness of the No. 15 coal in the study area using the PCA attributes as the inputs. Comparing the predicted results, it is noted that the trained ELM model with two selected PCA attributes yields better predication results than those from the other combinations of the attributes. Finally, the trained ELM model with real seismic data have a different number of hidden nodes (10) than the trained ELM model with synthetic seismic data. In summary, it is feasible to use an ELM model to predict the TDC thickness using the calculated PCA attributes as the inputs. However, the input attributes, the activation function and the number of hidden nodes in the ELM model should be selected and tested carefully based on individual application.

Hydrological flow rate estimation using artificial neural networks: Model development and potential applications

In present paper we develop a model of monthly river flow rate using artificial neural networks, based on the assumption that air temperature and precipitation predetermine the flow rate dynamics. In order to create a reliable prediction model, we used monthly observations made at Lim river basin in southwestern Serbia from 1950 to 2012, since results of this analysis could be significant for hydro-energy production at ``Potpeć" and ``Brodarevo" hydropower plants. Analysis was conducted using multilayer feed-forward perceptron with Levenberg–Marquardt learning algorithm and appropriate number of hidden neurons which provide the most reliable prediction accuracy. Analysis of derived models with different number of hidden nodes indicate that models are insensitive to the number of hidden units. Model with eight hidden nodes was chosen as the most reliable one, providing highest prediction accuracy (the highest values of determination coefficient and Nash–Sutcliffe coefficient). Predictive power of the developed model was tested against the recordings made in period 1991–2012, providing satisfying prediction accuracy. Moreover, Monte-Carlo simulation showed that prediction accuracy of developed models is robust against expected experimental error, confirming that derived models provide reliable predictions of flow rates, which could be used for water management plans and strategies. We also propose two potential applications of derived model: for predicting the future flow rate using the predefined climate models, and for forecasting the hydroenergy production, on the basis of the linear dependence of the observed flow rate and previously produced electric power. These application are verified for the regional climate model EBU-POM, for the period 2013–2100, and using the data on electric power production at hydro powerplant ``Potpeć".

Pornographic images recognition based on spatial pyramid partition and multi-instance ensemble learning

For tackling the problem of pornographic image recognition, a novel multi-instance learning (MIL) algorithm is proposed by using extreme learning machine (ELM) and classifiers ensemble. Firstly, a spatial pyramid partition-based (SPP) multi-instance modeling technique has been deployed to transform the pornographic images recognition problem into a typical MIL problem. The method has deployed a bag corresponding to an image and an instance corresponding to each partitioned sub-block described by low-level visual features (i.e. color, texture and shape). Secondly, a collection of visual word (VW) has been generated by using hierarchical k-mean clustering method, and then based on the fuzzy membership function between instance and VW, a fuzzy histogram fusion-based metadata calculation method has been proposed to convert each bag to a single sample, which allows the MIL problem to be solved directly by a standard single instance learning (SIL) machine. Finally, by using ELM, a group of base classifiers with different number of hidden nodes have been constructed, and their weights bas been dynamically determined by using performance weighting rule. Therefore, the strategy of classifiers ensemble is used to improve the overall adaptability of proposed ELMCE-MIL algorithm. Experimental results have shown that the method is robust, and its performance is superior to other similar algorithms.

Environmental impact of blasting at Drenovac limestone quarry (Serbia)

In present paper, the blast-induced ground motion and its effect on the neighboring structures are analyzed at the limestone quarry "Drenovac" in central part of Serbia. Ground motion is examined by means of existing conventional predictors, with scaled distance as a main influential parameter, which gave satisfying prediction accuracy (R > 0.8), except in the case of Ambraseys–Hendron predictor. In the next step of the analysis, a feed-forward three-layer back-propagation neural network is developed, with three input units (total charge, maximum charge per delay and distance from explosive charge to monitoring point) and only one output unit (peak particle velocity). The network is tested for the cases with different number of hidden nodes. The obtained results indicate that the model with six hidden nodes gives reasonable predictive precision (R ≈ 0.9), but with much lower values of mean-squared error in comparison to conventional predictors. In order to predict the influence level to the neighboring buildings, recorded peak particle velocities and frequency values were evaluated according to United States Bureau of Mines, USSR standard, German DIN4150, Australian standard, Indian DMGS circular 7 and Chinese safety regulations for blasting. Using the best conventional predictor, the relationship between the allowable amount of explosive and distance from explosive charge is determined for every vibration standard. Furthermore, the effect of air-blast overpressure is analyzed according to domestic regulations, with construction of a blasting chart for the permissible amount of explosive as a function of distance, for the allowable value of air-blast overpressure (200 Pa). The performed analysis indicates only small number of recordings above the upper allowable limit according to DIN4150 and DMGS standard, while, for all other vibration codes the registered values of ground velocity are within the permissible limits. As for the air-blast overpressure, no damage is expected to occur.

Applying Hierarchical Bayesian Neural Network in Failure Time Prediction

With the rapid technology development and improvement, the product failure time prediction becomes an even harder task because only few failures in the product life tests are recorded. The classical statistical model relies on the asymptotic theory and cannot guarantee that the estimator has the finite sample property. To solve this problem, we apply the hierarchical Bayesian neural network (HBNN) approach to predict the failure time and utilize the Gibbs sampler of Markov chain Monte Carlo (MCMC) to estimate model parameters. In this proposed method, the hierarchical structure is specified to study the heterogeneity among products. Engineers can use the heterogeneity estimates to identify the causes of the quality differences and further enhance the product quality. In order to demonstrate the effectiveness of the proposed hierarchical Bayesian neural network model, the prediction performance of the proposed model is evaluated using multiple performance measurement criteria. Sensitivity analysis of the proposed model is also conducted using different number of hidden nodes and training sample sizes. The result shows that HBNN can provide not only the predictive distribution but also the heterogeneous parameter estimates for each path.

Throughput analysis and bandwidth allocation for IEEE 802.11 WLAN with hidden terminals

Motivated by observations from real world wireless local area network (WLAN) deployments, we develop in this paper a novel analytical model to characterize the saturation throughput of an IEEE 802.11-based access point (AP) and stations under the influence of hidden terminals. Unlike existing models, our model can accommodate different numbers of hidden nodes without increasing the model complexity. Given any number of hidden nodes, only four constraints are needed to describe the interaction between stations and the AP with the consideration of both uplink and downlink traffic. Simulation evaluation shows that our model predicts network performance accurately over a wide range of network sizes and indicates the existence of a throughput starvation problem. To address this problem, based on our model, we formulate a bandwidth allocation problem to optimize the network throughput and fairness under some predefined requirements by systematically tuning the AP and stations contention windows. Simulation results show that the starvation problem is resolved with our approach, and the target throughput is met.

Hybrid neural network for prediction of CO2 solubility in monoethanolamine and diethanolamine solutions

The solubility of CO2 in single monoethanolamine (MEA) and diethanolamine (DEA) solutions was predicted by a model developed based on the Kent-Eisenberg model in combination with a neural network. The combination forms a hybrid neural network (HNN) model. Activation functions used in this work were purelin, logsig and tansig. After training, testing and validation utilizing different numbers of hidden nodes, it was found that a neural network with a 3-15-1 configuration provided the best model to predict the deviation value of the loading input. The accuracy of data predicted by the HNN model was determined over a wide range of temperatures (0 to 120 °C), equilibrium CO2 partial pressures (0.01 to 6,895 kPa) and solution concentrations (0.5 to 5.0M). The HNN model could be used to accurately predict CO2 solubility in alkanolamine solutions since the predicted CO2 loading values from the model were in good agreement with experimental data.

Automatic detection of clustered microcalcifications in digital mammograms using mathematical morphology and neural networks

In this paper we propose a new algorithm for the detection of clustered microcalcifications using mathematical morphology and artificial neural networks. Mathematical morphology provides tools for the extraction of microcalcifications even if the microcalcifications are located on a non-uniform background. Considering each mammogram as a topographic representation, each microcalcification appears as an elevation constituting a regional maximum. Morphological filters are applied, in order to remove: (a) noise and (b) regional maxima that do not correspond to calcifications. Each candidate object is marked as such, using a binary image. The original mammogram is used for the final feature extraction step. For the classification step we employ neural network classifiers. We review the performance of two multi-layer perceptrons (MLP) and two radial basis function neural networks (RBFNN) with different number of hidden nodes. The MLP with ten hidden nodes achieved the best classification score with a true positive detection rate of 94.7% and 0.27 false positives per image.

ESTIMATION OF NOX EMISSIONS IN THERMAL POWER PLANTS USING ENG-GENES NEURAL NETWORKS

This paper investigates neural network based estimation of NOx emissions in a thermal power plant, fed with both oil and methane fuels. Two types of neural network namely a novel ‘eng-genes' architecture and a Multilayer Perceptron (MLP) have been developed, both being optimised using genetic algorithms. Due to the local nature of the NOx generation process, operational information on the burner cells of the combustion chamber has been considered. Neural networks, with different numbers of hidden nodes have been tested on a set of three-dimensional data of the simulated combustion chamber. It is shown that, the proposed ‘eng-genes' neural network can produce accurate estimations with better generalisation performance than MLP.

Prediction of the continuous cooling transformation diagram of some selected steels using artificial neural networks

Continuous cooling transformation (CCT) diagrams play an important role in the description of the transformation behaviour of steels. The experimental determination of a CCT diagram is a very time consuming and expensive task. It would therefore be very attractive to be able to predict CCT diagrams from the chemical composition of the steel and its austenitising temperature. In this article the use of artificial neural networks for the prediction of the transformation start and finish lines in CCT diagrams is described. The data were selected from a single source: The vanadium steels, atlas of continuous cooling transformation diagrams [4].Three neural networks with different numbers of hidden nodes (5‐10‐15) were trained. The number of hidden nodes did not significantly influence the accuracy in the prediction. The network with the least number of hidden nodes (5) was therefore chosen for the evaluation of the performance of the neural networks. This neural network was able to predict the general trends in the CCT diagrams quite well. The relative standard deviation in the prediction of start and end temperatures of each transformation depended on the cooling rate. For the high and low cooling rates it was ∼ 40°C, for the intermediate it rose to 90°C for the ferrite start formation and to 75°C for the other diffusional transformations (pearlite and bainite).The accuracy of the predicted CCT diagram was primarily restricted by the modest quality of the input data used to train the neural network.

Prediction of the measured temperature after the last finishing stand using artificial neural networks

In this report the development of an artificial neural network, capable of predicting the temperature after the last finishing stand of a hot strip mill for a certain class of steels, is described. Three neural networks with different numbers of hidden nodes (3, 5 and 7) were trained. The relative standard deviation in finish temperature as predicted by the best performing neural network model (7 hidden nodes) was just over 25% smaller than that of the linear Hoogovens model. This improved accuracy can be explained by the incorrect assumption in the Hoogovens model of linear dependence of the finishing temperature on some input parameters.With the trained neural network, the influence of the various input parameters on the finishing temperature could be examined. The dependencies predicted by the neural network can be approximated by a linear fit and are a factor 2 lower for all input parameters.It is conceivable that operation of the mill using an artificial neural network for the prediction of the finishing temperature would have resulted in smaller operational fluctuations.

A neural network study on the dynamic identification of a fermentation system

The objective of this paper is to propose neural networks for the study of dynamic identification and prediction of a fermentation system which produces mainly 2,3-butanediol (2,3-BDL). The metabolic products of the fermentation, acetic acid, acetoin, ethanol, and 2,3-BDL were measured on-line via a mass spectrometer modified by the insertion of a dimethylvinylsilicone membrane probe. The measured data at different sampling times were included as the input and output nodes, at different learning batches, of the network. A fermentation system is usually nonlinear and dynamic in nature. Measured fermentation data obtained from the complex metabolic pathways are often difficult to be entirely included in a static process model, therefore, a dynamic model was suggested instead. In this work, neural networks were provided by a dynamic learning and prediction process that moved along the time sequence batchwise. In other words, a scheme of two-dimensional moving window (number of input nodes by the number of training data) was proposed for reading in new data while forgetting part of the old data. Proper size of the network including proper number of input/output nodes were determined by trained with the real-time fermentation data. Different number of hidden nodes under the consideration of both learning performance and computation efficiency were tested. The data size for each learning batch was determined. The performance of the learning factors such as the learning coefficient η and the momentum term coefficient α were also discussed. The effect of different dynamic learning intervals, with different starting points and the same ending point, both on the learning and prediction performance were studied. On the other hand, the effect of different dynamic learning intervals, with the same starting point and different ending points, was also investigated. The size of data sampling interval was also discussed. The performance from four different types of transfer functions, x/(1+|x|), sgn(x)·x 2/(1+x 2), 2/(1+e − x )−1, and 1/(1+e − x ) was compared. A scaling factor b was added to the transfer function and the effect of this factor on the learning was also evaluated. The prediction results from the time-delayed neural networks were also studied.

A learning algorithm for Recurrent Radial Basis Function Networks

This paper proposes a Recurrent Radial Basis Function network (RRBFN) that can be applied to temporal pattern classifications and predictions. Based on the architecture of the conventional Radial Basis Function networks, the RRBFNs use Gaussian nonlinearity and have feedback paths between every hidden node. These feedback paths enable the networks to learn temporal patterns without an input buffer to hold the recent elements of an input sequence. A gradient descent learning algorithm for the RRBFNs is derived. Two RRBFNs with different number of hidden nodes were tested using a temporal sequence generated by an infinite impulse response filter. The results show that the RRBFNs can approximate the filter more accurate than the Continually Running Fully Recurrent networks trained by the Real-Time Recurrent Learning algorithm.