Layer Artificial Neural Network Research Articles

Application of Multi Layer Artificial Neural Network in the Diagnosis System: A Systematic Review

Basic hardware comprehension of an artificial neural network (ANN), to a major scale depends on the proficientrealization of a distinctneuron. For hardware execution of NNs, mostly FPGA-designed reconfigurable computing systems are favorable .FPGA comprehension of ANNs through a hugeamount of neurons is mainlyan exigentassignment. This workconverses the reviews on various research articles of neural networks whose concernsfocused in execution of more than one input neuron and multilayer with or without linearity property by using FPGA. An execution technique through reserve substitution isprojected to adjust signed decimal facts. A detailed review of many research papers have been done for the <br /> proposed work.

Organic memristive devices for perceptron applications

One of the most challenging tasks in neuromorphic applications, in the field of artificial intelligence, is the hardware realization of artificial neural networks (ANNs) which are able to learn during information processing (pattern recognition and classification, approximation, prediction, etc).In this scenario, thanks to their ability to keep the memory of their previous conductive state, memristors are widely considered as promising elements for the efficient implementation of ANNs. In this paper we present a short review of the design and the hardware realization of single and double layer ANNs, which are able to perform linearly separable and non-separable logic classification, using organic memristive devices as elements, ensuring the network weight adjustment.

Multi-Scale Distributed Representation for Deep Learning and its Application to b-Jet Tagging

Recently machine learning algorithms based on deep layered artificial neural networks (DNNs) have been applied to a wide variety of high energy physics problems such as jet tagging or event classification. We explore a simple but effective preprocessing step which transforms each real-valued observational quantity or input feature into a binary number with a fixed number of digits. Each binary digit represents the quantity or magnitude in different scales. We have shown that this approach improves the performance of DNNs significantly for some specific tasks without any further complication in feature engineering. We apply this multi-scale distributed binary representation to deep learning on b-jet tagging using daughter particles' momenta and vertex information.

Artificial neural network and multiple regression analysis models to predict essential oil content of ajowan (Carum copticum L.)

Ajowan is an important medicinal plant that grows in arid and semi-arid regions of central Europe, India, Egypt, Iran, Iraq, Afghanistan, and Pakistan. Essential oil is the most consumable product of ajowan in pharmaceutical and food industrials, and correct predict of oil content is one of the main goals in breeding programs of ajowan. Two methods namely artificial neural network (ANN) and multiple regression model (MLR) were conducted to predict the oil content of ajowan from readily measurable plant characters. According to simple correlation analysis, four characters (number of rays, number of pedicels, number of flowers per umbellet, and number of umbellets in an umbel) were selected as input variables in both artificial neural network and multiple linear regressions models. The essential oil content of ajowan was well predicted using SigmoidAxon transfer function and two hidden layers of artificial neural network with a root mean square error (RMSE) of 0.192%, a mean absolute error (MAE) of 0.112% and a determination coefficient (R2) of 0.901. The performance of ANN was better than MLR with a RMSE of 0.262 and a R2 of 0.748. Based on stepwise regression and ANN analyses the most important characters for oil content of ajowan were number of umbellets in an umbel and number of flowers per umbellet and these traits can be assigned as selection criteria for essential oil content of ajowan.

A comparison between single layer and multilayer artificial neural networks in predicting diesel fuel properties using near infrared spectrum

ABSTRACTThe implementation of near infrared spectroscopy in monitoring diesel fuel properties is highly dependent on the capability of its predictive model. This study investigates the feasibility of a single layer artificial neural networks among various predictive models in predicting the diesel fuel properties using near infrared spectrum. Results were compared and discussed with that reported in previous studies that used the same data in predicting the diesel fuel properties. Findings show that the proposed single layer outperforms popular models, and is comparable with a recent advanced models in predicting the diesel fuel properties using near Infrared spectrum without data reduction.

Machine learning and artificial neural network prediction of interfacial thermal resistance between graphene and hexagonal boron nitride.

High-performance thermal interface materials (TIMs) have attracted persistent attention for the design and development of miniaturized nanoelectronic devices; however, a large number of potential new materials exist to form these heterostructures and the explorations of their thermal properties are time consuming and expensive. In this work, we train several supervised machine learning (ML) and artificial neural network (ANN) models to predict the interfacial thermal resistance (R) between graphene and hexagonal boron-nitride (hBN) with only the knowledge of the system temperature, coupling strength between two layers, and in-plane tensile strains. The training data were obtained by high-throughput computations (HTCs) of R using classical molecular dynamics (MD) simulations. Four different ML models, i.e., linear regression, polynomial regression, decision tree and random forest, are explored. A pair of one dense layer ANNs and another pair of two dense layer deep neural networks (DNNs) are also investigated. It is reported that the DNN models provide better R prediction results compared to the ML models. The thermal property predictions using HTC and ML/ANN models are applicable to a wide range of materials and open up new perspectives in the explorations of TIMs.

Flower image classification with basket of features and multi layered artificial neural networks

Artificial intelligence is penetrating most of the classification and recognition tasks performed by a computer. This work proposes to classify flower images based on features extracted during segmentation and after segmentation using multiple layered neural networks. The segmentation models used are watershed, wavelet, wavelet fusion model, supervised active contours based on shape, color and Local binary pattern textures and color, fused textures based active contours. Multi-dimension feature vectors are constructed from these segmented results for each indexed flower image labelled with their name. Each feature becomes input to a neuron in various feature layers and error back propagation algorithm with convex optimization structure trains these multiple feature layers. Testing with different flower images sets from multiple sources resulted in average classification accuracy of 92% for shape, color and texture supervised active contour segmented flower images.

Variable complexity online sequential extreme learning machine, with applications to streamflow prediction

In situations where new data arrive continually, online learning algorithms are computationally much less costly than batch learning ones in maintaining the model up-to-date. The extreme learning machine (ELM), a single hidden layer artificial neural network with random weights in the hidden layer, is solved by linear least squares, and has an online learning version, the online sequential ELM (OSELM). As more data become available during online learning, information on the longer time scale becomes available, so ideally the model complexity should be allowed to change, but the number of hidden nodes (HN) remains fixed in OSELM. A variable complexity VC-OSELM algorithm is proposed to dynamically add or remove HN in the OSELM, allowing the model complexity to vary automatically as online learning proceeds. The performance of VC-OSELM was compared with OSELM in daily streamflow predictions at two hydrological stations in British Columbia, Canada, with VC-OSELM significantly outperforming OSELM in mean absolute error, root mean squared error and Nash-Sutcliffe efficiency at both stations.

Machine Learning for Silver Nanoparticle Electron Transfer Property Prediction.

Nanoparticles exhibit diverse structural and morphological features that are often interconnected, making the correlation of structure/property relationships challenging. In this study a multi-structure/single-property relationship of silver nanoparticles is developed for the energy of Fermi level, which can be tuned to improve the transfer of electrons in a variety of applications. By combining different machine learning analytical algorithms, including k-mean, logistic regression, and random forest with electronic structure simulations, we find that the degree of twinning (characterized by the fraction of hexagonal closed packed atoms) and the population of the {111} facet (characterized by a surface coordination number of nine) are strongly correlated to the Fermi energy of silver nanoparticles. A concise three layer artificial neural network together with principal component analysis is built to predict this property, with reduced geometrical, structural, and topological features, making the method ideal for efficient and accurate high-throughput screening of large-scale virtual nanoparticle libraries and the creation of single-structure/single-property, multi-structure/single-property, and single-structure/multi-property relationships in the near future.

Application of artificial neural network for comparison and modeling of the ultrasonic and stirrer assisted removal of anionic dye using activated carbon supported with nanostructure material

In this study, a green approach has been described for the synthesis of copper sulfide nanoparticles loaded on activated carbon (CuS‐NP‐AC) and usability of it for the removal of sunset yellow (SY) dye by ultrasound‐assisted and stirrer has been compared. In addition, the artificial neural network (ANN) model has been employed for a forecasting removal percentage of SY dye using the results obtained. This material was characterized using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The impact of variables, including initial dye concentration (mg/L), pH, adsorbent dosage (g), sonication time (min) and temperature (°C) on SY removal was studied. Fitting the experimental equilibrium data of different isotherm models such as Langmuir, Freundlich, Temkin and Dubinin–Radushkevich models display the suitability and applicability of the Langmuir model. Analysis of experimental adsorption data of different kinetic models including pseudo‐first and second order, Elovich and intraparticle diffusion models indicate the applicability of the second‐order equation model. The adsorbent (0.005 g) is applicable for successful removal of SY dye (> 98%) in short time (9 min) under ultrasound condition. A three layer ANN models with 8 and 6 neurons at hidden layer was selected as optimal models using stirrer and ultrasonic, respectively. These models displayed a good agreement between forecasted data and experimental data with the determination coefficient (R2) of 0.9948 and 0.9907 and mean squared error (MSE) of 0.0001 and 0.0002 for training set using stirrer and ultrasonic, respectively.

The neural network as a part of decision support system for quality management for production objects in machining process

The research discusses the use of artificial neural networks (ANN) as components of a decision support system (DSS) to automate quality control manufacturing facilities machining business at the production, which should be focused on the analysis of large amounts of heterogeneous information. The necessity to use ANN as a part of DSS is justified by the fact that quality control during production is multistage and time-consuming process that is formalized difficult, and moreover requires considerable information and material costs for the efficiency of manufacturing operations performed. Taking into account the existing experience of successful use of ANN to solve difficult formal problems associated with handling large volumes of diverse and rapidly changing information, the authors synthesized ANN for automated determination of the causes deterioration of the quality of production objects (PO) in the performance of manufacturing operations application. Particular attention is paid to the definition of the dimension of the hidden layer ANN synthesized due to the fact that today still there is no analytical expression to determine the dimension of the hidden layer ANN and size of the latter is determined only by the experimental results of ANN several different structures by comparison the results, in particular the value of mean square error.

Comparative analysis on hidden neurons estimation in multi layer perceptron neural networks for wind speed forecasting

In this paper methodologies are proposed to estimate the number of hidden neurons that are to be placed numbers in the hidden layer of artificial neural networks (ANN) and certain new criteria are evolved for fixing this hidden neuron in multilayer perceptron neural networks. On the computation of the number of hidden neurons, the developed neural network model is applied for wind speed forecasting application. There is a possibility of over fitting or under fitting occurrence due to the random selection of hidden neurons in ANN model and this is addressed in this paper. Contribution is done in developing various 151 different criteria and the evolved criteria are tested for their validity employing various statistical error means. Simulation results prove that the proposed methodology minimized the computational error and enhanced the prediction accuracy. Convergence theorem is employed over the developed criterion to validate its applicability for fixing the number of hidden neurons. To evaluate the effectiveness of the proposed approach simulations were carried out on collected real-time wind data. Simulated results confirm that with minimum errors the presented approach can be utilized for wind speed forecasting. Comparative analysis has been performed for the estimation of the number of hidden neurons in multilayer perceptron neural networks. The presented approach is compact, enhances the accuracy rate with reduced error and faster convergence.

A New Multilevel Input Layer Artificial Neural Network for Predicting Flight Delays at JFK Airport

One of the biggest problems for major airline is predicting flight delay. Airlines try to reduce delays to gain the loyalty of their customers. Hence, a prediction model that airliners can use to forecast possible delays is of significant importance. In this regards, artificial neural network (ANN) techniques can be beneficial for this application. One of the main challenges of using ANNs is handling nominal variables. 1-of-N encoding is widely used to deal with this problem, however, this method is known to reduce the performance of ANN's by introducing multicollinearity. In this paper, we introduce a new type of multilevel input layer ANN that can handle nominal variables and is interpretable in a sense that one can easily see the relationships between different input variables and output variables. As a case study, the proposed method was applied to predict the delay of incoming flights at JFK airport, where the neurons of each sublayer of the input layer symbolize the delay sources at different levels of the system, and the activation of each neuron represents the possibility of being the source of overall delay. Finally, we compared the proposed approach with the traditional gradient descent back propagation ANN model and the proposed model was able to outperform the traditional backpropagation method in terms of the prediction error (root mean squared error) and time required to train the ANN model.

An Artificial Neural Network Model for Predicting the Pressure Gradient in Horizontal Oil–Water Separated Flow

In this study, a three–layer artificial neural network (ANN) model was developed to predict the pressure gradient in horizontal liquid–liquid separated flow. A total of 455 data points were collected from 13 data sources to develop the ANN model. Superficial velocities, viscosity ratio and density ratio of oil to water, and roughness and inner diameter of pipe were used as input parameters of the network while corresponding pressure gradient was selected as its output. A tansig and a linear function were chosen as transfer functions for hidden and output layers, respectively and Levenberg–Marquardt back–propagation algorithm were applied to train the ANN. The optimal topology of the ANN was achieved with 16 neurons in hidden layer, which made it possible to estimate the pressure gradient with a good accuracy (R2=0.996 &AAPE=7.54%). In addition, the results of the developed ANN model were compared to Al–Wahaibi correlation results (with R2=0.884&AAPE=17.17%) and it is found that the proposed ANN model has higher accuracy. Finally, a sensitivity analysis was carried out to investigate the relative importance of each input parameter on the ANN output. The results revealed that the pipe diameter (D) has the most relative importance (24.43%) on the ANN output, while the importance of the other parameters is nearly the same.

Detecting and ranking cash flow risk factors via artificial neural networks technique

Jensen (1986) predicts that managers with more free cash flows (FCF) available show an opportunistic behavior, which leads to the emergence of the FCF risk. Consequently, determining the relative importance of factors affecting the FCF risks is vital. The major purpose of this article is testing Jensen's assertion by identifying relative significance of the factors influencing FCF risks via artificial neural network (ANN). In effect, seven independent variables relating to FCF risks including debt policy, ownership concentration, ownership level, managerial ownership, state ownership, size of the firm, and profitability draw on the literature. The study collects 1224 company-year data from Tehran Stock Exchange (TSE) for the period 2001–2010. The study applies the Pearson's correlation and a three layer ANN. The results of testing the hypotheses show that, among the preceding 7 factors, profitability is the most important factor in determining FCF risks. Following profitability, the second and third most important factors are the debt policy and size of the company. Consequently, this finding posits a great implication for the stock markets and contradicts Jensen's (1986) study.

Multi-objective optimization of cut quality characteristic in CO2 laser cutting stainless steel

Original scientific paper In this paper, multi-objective optimization of the cut quality characteristics in CO2 laser cutting of AISI 304 stainless steel was discussed. Three mathematical models for the prediction of cut quality characteristics such as surface roughness, kerf width and heat affected zone were developed using the artificial neural networks (ANNs). The laser cutting experiment was planned and conducted according to the Taguchi’s L27 orthogonal array and the experimental data were used to train single hidden layer ANNs using the Levenberg-Marquardt algorithm. The ANN mathematical models were developed considering laser power, cutting speed, assist gas pressure, and focus position as the input parameters. Multi-objective optimization problem was formulated using the weighting sum method in which the weighting factors that are used to combine cut quality characteristics into the single objective function were determined using the analytic hierarchy process method.

Low Power Adder based ANN

This paper presents an overview of datapath realizations of the Hardware neural network models which perform massive parallel operations for best results and real time applications. Digital implemented neural models processing element – adder with low power consumption is proposed for real-time multimedia applications. Proposed adder is illustrated in the 23-1 tree layer artificial neural network (ANN). Designs were modeled with Verilog HDL and implemented in FPGA domain by targeting the Virtex 7 device.

Application of Soft Computing Techniques for Modeling and Optimization of Slotted-Electrical Discharge Diamond Face Grinding Process

An innovative hybrid machining called as slotted-electrical discharge diamond face grinding (S-EDDFG) has been developed by comprising the electrical discharge face grinding and diamond face grinding. Due to complexity in S-EDDFG process, the selection of suitable combinations of input parameters is a challenging task for manufacturers. In recent, soft computing techniques are gaining more attention of researchers for modeling and optimization for the process parameters of complex manufacturing processes. In the present paper, two soft computing techniques known as artificial neural network (ANN) and non-dominated sorting genetic algorithm-II (NSGA-II) have been applied for modeling and optimization of process parameters (pulse current, pulse on-time, pulse off-time, wheel RPM and grit number) during machining of Aluminium–Silicon Carbide–Graphite (Al/SiC/Gr) composite. The results shown that two hidden layer ANN architecture was more appropriate to predict the material removal rate and average surface roughness during S-EDDFG of Al/SiC/Gr workpiece. The optimal results obtained by combined approach of ANN–NSGA-II are experimentally validated and found that optimal data are under acceptable limits.

Design of a feature-tuned ANN model based on bulk rock-derived mineral spectra for endmember classification of a hyperspectral image from an iron ore deposit

In this article, a new feature-tuned artificial neural network (ANN) model has been developed for endmember classification of a hyperspectral image. This model is developed on the basis of using only the essential absorption bands of mineral spectra as opposed to using all the spectral bands of the hyperspectral image. This approach has the added advantages of reducing the dimensionality of input features to the ANN as well as inhibiting the influences of noisy bands for classification of endmembers. The proposed ANN model is trained using input features extracted from laboratory spectra of in situ bulk ore materials collected from an existing iron ore deposit. The input features are basically the constituent absorption bands of mineral spectra where each absorption band is mathematically characterized by the centre, width, and strength parameters of a Gaussian curve. For extracting absorption bands from a mineral spectrum, a modified Gaussian model has been used. The application of this model also necessitates the design of a special template for the input layer ANN model. After training the model, its generalization property is assessed through a testing data set. The model has achieved nearly 97% of classification accuracy in a training set, and 71% of accuracy in a testing set. The trained model is then applied on Hyperion imagery collected over an iron ore deposit. All the endmember spectra of this deposit are classified into either vegetation or any of the ores or rock present in the deposit. None of the endmembers is classified into non-iron ore minerals.

Application of artificial neural network for predicting crack growth direction in multiple cracks geometry

The objective of this study is to design an efficient artificial neural network (ANN) architecture in order to predict the crack growth direction in multiple crack geometry. Nonlinear logistic (sigmoid and tangent hyperbolic) and linear activation functions have been used through the one- and two-hidden layer ANN. 85 tests were conducted on aluminium alloys under different crack positions, defined by crack tip distance, crack offset distance, crack size, and crack inclination with loading axis. The experimental data set as first degree or second degree were used to train 22 proposed ANN models to predict the output for new data sets (not included in the training sets). The model results were then compared with the experimental data. It was observed that ANN model with combinations of activation functions and two hidden layers predict the crack initiation direction with good accuracy when higher order input variables are presented to the network.