Back-propagation Multilayer Perceptron Research Articles

Predicting the helpfulness of online reviews using multilayer perceptron neural networks

With the great development of e-commerce, users can create and publish a wealth of product information through electronic communities. It is difficult, however, for manufacturers to discover the best reviews and to determine the true underlying quality of a product due to the sheer volume of reviews available for a single product. The goal of this paper is to develop models for predicting the helpfulness of reviews, providing a tool that finds the most helpful reviews of a given product. This study intends to propose HPNN (a helpfulness prediction model using a neural network), which uses a back-propagation multilayer perceptron neural network (BPN) model to predict the level of review helpfulness using the determinants of product data, the review characteristics, and the textual characteristics of reviews. The prediction accuracy of HPNN was better than that of a linear regression analysis in terms of the mean-squared error. HPNN can suggest better determinants which have a greater effect on the degree of helpfulness. The results of this study will identify helpful online reviews and will effectively assist in the design of review sites.

Artificial Neural Network prediction of Cu–Al2O3 composite properties prepared by powder metallurgy method

Artificial Neural Networks (ANNs) are excellent tools for prediction of complex processes that have many variables and complex interactions. In the present study, the properties of copper based composite prepared from sintering of mechanically alloyed powders, were predicted using Artificial Neural Network (ANN) approach. In order to prepare copper based composites, copper powder with four different amounts of Al2O3 reinforcement (1, 1.5, 2, 2.5wt%) were mechanically alloyed and the consolidated compacts of prepared powders were sintered in five different temperatures of 725–925°C at seven several sintering times of 15–180min. Hardness and electrical conductivity measurements were performed to evaluate the properties of these composites. Then, for modeling and prediction of hardness and electrical conductivity, a multi layer perceptron back propagation feed forward neural network was constructed to evaluate and compare the experimental calculated data to predicted values. It was found that, in a given sintering temperature of 875°C, the electrical conductivity increases as the sintering time increases and the amount of Al2O3 reinforcement decreases. Also, increasing of reinforcement amount and sintering time in a given sintering temperature of 875°C leads to a decrease in hardness. Furthermore, electrical conductivity and hardness of specimens have shown a consistency with predicted results of ANN. These trained values had an average error of 3% and 5% for electrical conductivity and hardness values, respectively.

Neural Network Based Resonant Frequency Solver for Rectangular‐Shaped Shorting Pin‐Loaded Antennas

ABSTRACTThis study presents an artificial neural network (ANN) estimation of the operating frequencies of shorting pin‐loaded rectangular microstrip patch antennas. A feed forward back propagation multilayer perceptron neural network structure is applied in the study. The results are compared with the ones in the literature and the FEM based simulation results. The results of the operating frequencies obtained by using this method are in very good agreement with the experimental results presented in the literature. Several antennas are also simulated by a finite element method based solver and these results are also compared with the results of the proposed neural network model. The average error of the lower frequency obtained by this study has a decrement of 2.025% when compared to the FEM based simulation software and for the upper frequency this difference is 6.835%. The effects of permittivity of the antenna, size of the dimensions of the rectangular patch, and the shorting pin position are also evaluated. In the light of the ANN model and the relations obtained two antennas in the same shape are produced and the results of these antennas are presented as well. © 2013 Wiley Periodicals, Inc. Microwave Opt Technol Lett 55:3025–3028, 2013

Combining neural networks and ANFIS classifiers for supervised examining of electrocardiogram beats

In this paper, the supervised classification of the electrocardiogram (ECG) beats based on the fusion of several intelligent learning machines is described. For classification of ECG heartbeats, first, the QRS complexes are delineated by an efficient algorithm so as to identify the fiducial and J-locations of each complex. For each delineated QRS complex, a feature vector is established based on the geometrical properties of the complex waveform and its associated discrete-wavelet transform. Next, three different multi-layer perceptron back-propagation (MLP-BP) networks are trained with different topologies and intrinsic parameters. Afterwards, the outputs of MLP-BPs are used as the new feature space elements for training three adaptive fuzzy network inference systems (ANFIS) in order to increase the final accuracy. At the end, the outputs of ANFIS classifiers are voted based on majority for each input sample. The method was applied to seven arrhythmias (Normal, LBBB, RBBB, PVC, APB, VE, VF) which belong to the MIT-BIH Arrhythmia Database and the average accuracy value Acc = 98.28% was achieved for the beat-level. Also, the proposed method was assessed to five arrhythmias (Normal, LBBB, RBBB, PVC, APB) according to validation standards of the American Heart Association (AHA) at record (subject) level and the average accuracy value Acc = 73.39% was achieved. To evaluate performance quality of the new proposed hybrid learning machine, the obtained results were compared with similar peer-reviewed studies in this area.

Text Dependent Speaker Recognition using MFCC features and BPANN

Mel-Frequency Cepstral Coefficients are spectral feature which are widely used for speaker recognition and text dependent speaker recognition systems are the most accurate in voice based authentication systems. In this paper, a text dependent speaker recognition method is developed. MFCCs are computed for a selected sentence. The first 13 MFCCs are considered for each frames of duration 26ms and each coefficient is clustered to a 5 element cluster centres and finally to a form a 65 element speech code vector for the entire speech. The speech code is trained using a multi-layer perceptron backpropagation gradient descent network and the network is tested for various test patterns. The performance is measured using FAR, FRR and EER parameters. The recognition rate achieved is 96.18% for a cluster size of 5 in each coefficient. General Terms Speaker Recognition, Artificial Neural Networks, Clustering

Neural network simulation of spring flow in karst environments

Daily discharges of two springs lying in a karstic environment were simulated for a period of 2.5 years with the use of a multi-layer perceptron back-propagation neural network. Two models were developed for the springs, one relying on the original data and another where the missing discharge values were supplemented by assuming linear relationships during base flow conditions. For both springs the mean square error of the two models did not differ significantly, with an improvement exhibited at the extremes, during the network’s training phase, by the model that utilized the extended data set, the results of which are reported here. The time lag between precipitation and spring discharge differed significantly for the two springs indicating that in karstic environments hydraulic behavior is dominated, even within a few hundred meters, by local conditions. Optimum training results were attained with a Levenberg–Marquardt algorithm resulting in a network architecture consisting of two input layer neurons, four hidden layer neurons, and one output layer neuron, the spring’s discharge. The neural network’s predictions captured the behavior for both springs and followed very closely the discontinuities in the discharge time series. Under-/over-estimation of observed discharges for the two springs remained below 3 %, with the exception of a few local maxima where the predicted discharges diverged more strongly from observed values. Inclusion of temperature data did not add to the improvement of predictions. Finally, optimum predictions were attained when past discharge data were added to the input record and discharge differentials rather than direct discharges were calculated resulting in elimination of any local maximum discrepancy between observed and predicted discharge values.

Identification of monofloral honey using voltammetric electronic tongue

Quality assessment of honey is often related to its floral origin which is a complex task to evaluate. Traditional technique of floral assessment is made by melissopalynological method. However, this method is quite time consuming and also often operator dependent. Thus, the fallout is a large range of error in interpretation of the result and hence there is considerable demand for instrumental methods to assess the identification of pollen in honey. In this pursuit, an electronic tongue based on cyclic voltammetry is developed to discriminate honey samples based on their floral types and is described in this paper. The technique has been investigated using platinum as the working electrode and the resultant current from the potentiostat has been considered for data analysis. The use of principal component analysis (PCA) and linear discriminate analysis (LDA) proves to be useful in clustering honey samples. Finally, classification performances are investigated using back-propagation multilayer perceptron (BP-MLP) and radial basis function (RBF) neural network models for identification of different floral origin of honey.

ANS-based preprocessing of company performance indicators

This work examines an artificial neural system (ANS) capable of dimensionality-reduction and its fitness to a business data analysis problem. While ANSs are often used in the later stages of explorative data analysis to place similar cases in clusters or to identify known patterns, the role of the ANS here is in the early pre-processing stage, to produce a smaller number of new variables which maintain important structures in the data. The ANS examined is a specially configured variation of the back-propagation multilayer perceptron (BP-MLP). The method is tested against others, on a supervised classification problem where the goal is to separate groups of similarly performing companies in the food and beverage sector of the Greek industry from their economic and non-economic characteristics. A typical pattern recognition algorithm is applied to classify the companies in performance groups, using a minimal pre-processing of the original company data b dimensionality-reduction through principal component analysis (PCA) c variables resulting from the examined ANS-based pre-processing method. The method’s ability to produce useful variables is evaluated by comparing the various groups of companies produced by the classifier algorithm using the three procedures.

Predicting concrete properties using neural networks (NN) with principal component analysis (PCA) technique

This paper discusses the combined application of two different techniques, Neural Networks (NN) and Principal Component Analysis (PCA), for improved prediction of concrete properties. The combination of these approaches allowed the development of six neural networks models for predicting slump and compressive strength of concrete with mineral additives such as blast furnace slag, fly ash and silica fume. The Back-Propagation Multi-Layer Perceptron (BPMLP) with Bayesian regularization was used in all these models. They are produced to implement the complex nonlinear relationship between the inputs and the output of the network. They are also established through the incorporation of a huge experimental database on concrete organized in the form Mix-Property. Thus, the data comprising the concrete mixtures are much correlated to each others. The PCA is proposed for the compression and the elimination of the correlation between these data. After applying the PCA, the uncorrelated data were used to train the six models. The predictive results of these models were compared with the actual experimental trials. The results showed that the elimination of the correlation between the input parameters using PCA improved the predictive generalisation performance models with smaller architectures and dimensionality reduction. This study showed also that using the developed models for numerical investigations on the parameters affecting the properties of concrete is promising.

A genetic algorithm model based on artificial neural network for prediction of the axillary lymph node status in breastcancer

Axillary Lymph Node (ALN) status is an extremely important factor to assess metastatic breast cancer. Surgical operations which may be necessary and cause some adverse effects are performed in determination ALN status. The purpose of this study is to predict ALN status by means of selecting breast cancer patient's basic clinical and histological feature(s)that can be obtained in each hospital. 270 breast cancer patients' data are collected from Ankara Numune Educational and Research Hospital and Ankara Oncology Educational and Research Hospital. These are classified using back propagation MultiLayer Perceptron (MLP), Logistic Regression (LR) and Genetic Algorithm (GA) based MLP models. Receiver Operating Characteristics (ROC) such as sensitivity, specificity, accuracy and area under of ROC (AUC) and regression are used to evaluate performances of the developed models. It is concluded from LR and GA based MLP, that menopause status and lymphatic invasion are the most significant features for determining ALN status. GA provides to select best features as MLP inputs. It also optimizes the weights of backpropagation algorithm in MLP. The values of regression and accuracy of the GA based MLP with 9features (numerical age, categorical age, menopause status, tumor size, tumor type, tumor location, T staging, tumor grade and lymphatic invasion) are found as 0.96 and 98.0% with respectively. According to results, proposed GA based MLP classifier can be used to predict the ALN status of breast cancer without surgical operations.

Analysis of micrornas by neural network for early detection of cancer

Problem Statement: Cancer is a complex genetic disease. Evidence is emerging that particular microRNAs may play a role in human cancer pathogenesis; they exhibit important regulatory roles in development, cell proliferation, cell survival and apoptosis [1]. The effectiveness in treatment and curing cancer is directly dependent on the ability to detect cancers at their earlier stages. This study developed a neural network with back-propagation learning algorithm for the prediction of microRNAs responsible from cancer pathogenesis at earlier stages. Purpose of the Study: The aim of this research was to develop a nonlinear empirical model to predict a score value for specific microRNAs responsible from cancer pathogenesis by using micro-array data. Methods: The Back-Propagation Multi Layer perceptron (BPMLP) for the prediction of score value. The supervised learning algoritm is used during the training and testing stages. The number of inputs is fixed at two and the number of output is one. The 677 micro-array data were collected from a cancer microarray database and data-mining platform called Oncomine. Testing data are classified into randomly selected four groups of data. Findings and ResultsIn this work, a BPMLP, is used to establish the relationship between the microRNAs, p-value and score value. Accuracy rate was adjusted to 80% and 90%. The best performance was obtained with 20 neurons in the hidden layer and learning parameter as 0.00099. Conclusions and Recommendations: BPMLP is obtained by developing forward propagation phase and sensitivity network. It is shown that, the generalization capability can be significantly improved by optimizing some of the parameters such as learning parameter, and neuron number. Since microRNAs may play a role in human cancer pathogenesis, we have shown that, it is possible to predict relationships of microRNAs, p-values with different score values. Therefore, by using BPMLP, it is possible to predict microRNAs responsible from cancer pathogenesis at earlier stages.

Prediction of Natural Frequency of Laminated Composite Plates Using Artificial Neural Networks

The paper is focused on the application of artificial neural networks (ANN) in predicting the natural frequency of laminated composite plates under clamped boundary condition. For training and testing of the ANN model, a number of finite element analyses have been carried out using D-optimal design in the design of experiments (DOE) by varying the fibre orientations, –45?, 0?, 45? and 90?. The composite plate is modeled using linear layered structural shell element. The natural frequencies were found by analyses which were done by finite element (FE) analysis software. The ANN model has been developed using multilayer perceptron (MLP) back propagation algorithm. The adequacy of the developed model is verified by coefficient of determination (R). It was found that the R2 (R: coefficient of determination) values are 1 and 0.998 for train and test data respectively. The results showed that, the training algorithm of back propagation was sufficient enough in predicting the natural frequency of laminated composite plates. To judge the ability and efficiency of the developed ANN model, absolute relative error has been used. The results predicted by ANN are in very good agreement with the finite element (FE) results. Consequently, the D-optimal design and ANN are shown to be effective in predicting the natural frequency of laminated composite plates.

Non-parametric modelling of a rectangular flexible plate structure

This research investigates the performance of dynamic modelling using non-parametric techniques for identification of a flexible structure system for development of active vibration control. In this paper, the implementation details are described and the experimental studies conducted in this research are analysed. The input–output data of the system were first acquired through the experimental studies using National Instruments (NI) data acquisition system. A sinusoidal force was applied to excite the flexible plate and the dynamic response of the system was then investigated. Non-parametric modelling of the system were developed using several artificial intelligent methodologies namely Adaptive Elman Neural Networks (ENN), Backpropagation Multi-layer Perceptron Neural Networks (MLPNN) and Adaptive Neuro-Fuzzy Inference System (ANFIS). The performance of all these methodologies were compared and discussed. Finally, validation and verification of the obtained model was conducted using One Step Ahead (OSA) prediction, mean squared error (MSE) and correlation tests. The prediction ability of the model was further observed with unseen data. The results verified that the MLPNN converge to an optimum solution faster and the dynamic model obtained described the flexible plate structure very well. The non-parametric models of the flexible plate structure thus developed and validated will be used as the representation of the transfer function of the system in subsequent investigations for the development of active vibration control strategies for vibration suppression in flexible structures.

Gases concentration estimation using heuristics and bio-inspired optimization models for experimental chemical electronic nose

By virtue of an electronic nose, detection and concentration estimation of harmful gases indoor become feasible by using a multi-sensor system. The estimation accuracy in actual application is constantly aspired by manufactures and researchers. This paper analyzes the application of different bio-inspired and heuristic techniques to the problem of concentration estimation in experimental electronic nose application. In this paper, seven different particle swarm optimization models are considered including six models used before in numerical function optimization, and a novel hybrid model of particle swarm optimization and adaptive genetic algorithm, for optimizing back-propagation multilayer perceptron neural network. We describe the performance of a particle swarm optimization technique, an adaptive genetic strategy and a back-propagation artificial neural network approach to perform concentration estimation of chemical gases and improve the intelligence of an E-nose.

Rainfall disaggregation in non-recording gauge stations using space-time information system

The disaggregation of coarser Precipitation data will help to adjust the deficit of unavailability of data in non-recording gauge stations. The Artificial Neural Network (ANN) facilitates to adjust the rainfall time steps into desired small scales. At first, the Geostatistical method of co-kriging was used for mapping purposes to find the missing duration and depth of rainfall of some incomplete data stations in Sydney Australia. Then, since there was no information about the breakpoint data in non-recording target central station 7261, a process was performed to disaggregate the data of recording gauge station sited besides this non-recording one. Definitely, a similar station was delineated, firstly Thiessen polygon was used instead of station 7261 and then the results of applying two different ANN models (a feed forward back propagation multilayer perceptron (MLP) and a Radial Basis Function (RBF) network) were evaluated to disaggregate the data of this station, and the best disaggregation model was introduced.

Prediction of the vortex yarn tenacity from some process and nozzle parameters based on numerical simulation and artificial neural network

In this study, two methods, the numerical simulation and the artificial neural network, are adopted to predict the vortex yarn tenacity from some process and nozzle parameters. The fiber-airflow interaction and the motional characteristics of the fiber inside the vortex spinning nozzle are analyzed through the numerical simulation method to predict the yarn tenacity from the parameters of the jet orifice diameter and the yarn delivery speed. The back-propagation multilayer perceptron (MLP) neural network is adopted to predict the tenacity of vortex yarn spun from modal fibers. The following parameters are used as inputs: nozzle pressure, jet orifice angle, distance between the guiding needle and the spindle, and spindle cone angle. The predicted and measured tenacities show low relative errors and a high correlation coefficient, indicating the artificial neural network method provides accurate prediction results. In addition, the effects of these parameters on the yarn tenacity are also analyzed.

Application of artificial neural networks for simulation of experimental CO 2 absorption data in a packed column

The generalization performances of the Back Propagation Multi-Layer Perceptron (BPMLP) and the Radial Basis Function (RBF) neural networks were compared together by resorting to several sets of experimental data collected from a pilot scale packed absorption column. The experimental data were obtained from an 11 cm diameter packed tower filled with 1.8 m ¼ inch ceramic Rashig rings. The column was used for separation of carbon dioxide from air using various concentrations and flow rates of Di-Ethanol Amine (DEA) and Methyl Di-Ethanol Amine (MDEA) solutions. Two in-house efficient algorithms were employed for optimal training of both neural networks. The simulation results indicated that the RBF networks can perform more adequately than the MLP networks for filtering the noise (measurement errors) and capturing the true underlying trend which is essential for a reliable generalization performance.

Synthesis of multiple-type classification algorithms for robust heart rhythm type recognition: Neuro-svm-pnn learning machine with virtual QRS image-based geometrical features

The paper addresses a new QRS complex, geometrical feature extraction technique, as well as its application in supervised electrocardiogram (ECG) heart-beat hybrid (fusion) classification. To this end, after detection and delineation of the major events of an ECG signal via an appropriate algorithm, each QRS region and also its corresponding Discrete Wavelet Transform (DWT) are supposed as virtual images, and each one is divided into eight polar sectors. Then, the curve length of each excerpted segment is calculated and used as an element of the feature space. To increase the robustness of the proposed classification algorithm versus noise, artifacts and arrhythmic outliers, a fusion structure consisting of four different classifiers, namely Support Vector Machine (SVM), Probabilistic Neural Network (PNN) and two Multi Layer Perceptron-Back Propagation (MLP-BP), with different topologies, were designed. To show the merit of the new proposed algorithm, it was applied to all MIT-BIH arrhythmia database records, and the discriminative power of the classifier in isolation of different beat types of each record was assessed. As a result, the average accuracy value, Acc=98.18%, was obtained. Also, the proposed method was applied to 8 arrhythmias and an average value of Acc=97.37% was achieved.

Development and Implementation of Parameterized FPGA-Based General Purpose Neural Networks for Online Applications

This paper presents the development and implementation of a generalized backpropagation multilayer perceptron (MLP) architecture described in VLSI hardware description language (VHDL). The development of hardware platforms has been complicated by the high hardware cost and quantity of the arithmetic operations required in online artificial neural networks (ANNs), i.e., general purpose ANNs with learning capability. Besides, there remains a dearth of hardware platforms for design space exploration, fast prototyping, and testing of these networks. Our general purpose architecture seeks to fill that gap and at the same time serve as a tool to gain a better understanding of issues unique to ANNs implemented in hardware, particularly using field programmable gate array (FPGA). The challenge is thus to find an architecture that minimizes hardware costs, while maximizing performance, accuracy, and parameterization. This work describes a platform that offers a high degree of parameterization, while maintaining generalized network design with performance comparable to other hardware-based MLP implementations. Application of the hardware implementation of ANN with backpropagation learning algorithm for a realistic application is also presented.

Controlling and improving quality of the fertiliser production process using neural network models

Fertiliser production process is characterised by being a dynamic process which is not easy to be predicted and controlled due to uncertain, imprecise and vague parameters' relations. Although mathematical modelling techniques are very well developed, these types of dynamic processes are difficult to be modelled by those techniques and also the regression models are complex to be used for real time control and, usually, their errors are significant. The main and most important quality characteristic in the fertiliser production process is the moisture content. This parameter affects the product shelf life, effectiveness and harmful internal reactions. In this research, two different artificial neural network (ANN) approaches are developed to predict the moisture content of the produced fertiliser: the back-propagation multilayer perceptron (BPMLP) and the radial basic function (RBF) nets. The two models performed satisfactory in predicting the moisture content with low error percent. Predicting the moisture content, the quality of the produced fertiliser can be enhanced either by reheating, adding chemicals, or both.