Neural Networks in Bankruptcy Prediction - A Comparative Study on the Basis of the First Hungarian Bankruptcy Model

The article attempts to answer the question whether or not the latest bankruptcy prediction techniques are more reliable than traditional mathematical–statistical ones in Hungary. Simulation experiments carried out on the database of the first Hungarian bankruptcy prediction model clearly prove that bankruptcy models built using artificial neural networks have higher classification accuracy than models created in the 1990s based on discriminant analysis and logistic regression analysis. The article presents the main results, analyses the reasons for the differences and presents constructive proposals concerning the further development of Hungarian bankruptcy prediction.

Flexible pavement layer moduli determination : an adaptive artificial neural network approach

The estimation of pavement layer moduli through the use of an artificial neural network is a new concept which provides a less strenuous strategy for backcalculation procedures. Artificial Neural Networks are biologically inspired models of the human nervous system. They are specifically designed to carry out a mapping characteristic. This study demonstrates how an artificial neural network uses non-destructive pavement test data in determining flexible pavement layer moduli. The input parameters include plate loadings, corresponding sensor deflections, temperature of pavement surface, pavement layer thicknesses and independently deduced pavement layer moduli.

Critério de correntropia no treinamento de redes fuzzy wavelet neural networks para identificação de sistemas dinâmicos não lineares

The great interest in nonlinear system identification is mainly due to the fact that a large amount of real systems are complex and need to have their nonlinearities considered so that their models can be successfully used in applications of control, prediction, inference, among others. This work evaluates the application of Fuzzy Wavelet Neural Networks (FWNN) to identify nonlinear dynamical systems subjected to noise and outliers. Generally, these elements cause negative effects on the identification procedure, resulting in erroneous interpretations regarding the dynamical behavior of the system. The FWNN combines in a single structure the ability to deal with uncertainties of fuzzy logic, the multiresolution characteristics of wavelet theory and learning and generalization abilities of the artificial neural networks. Usually, the learning procedure of these neural networks is realized by a gradient based method, which uses the mean squared error as its cost function. This work proposes the replacement of this traditional function by an Information Theoretic Learning similarity measure, called correntropy. With the use of this similarity measure, higher order statistics can be considered during the FWNN training process. For this reason, this measure is more suitable for non-Gaussian error distributions and makes the training less sensitive to the presence of outliers. In order to evaluate this replacement, FWNN models are obtained in two identification case studies: a real nonlinear system, consisting of a multisection tank, and a simulated system based on a model of the human knee joint. The results demonstrate that the application of correntropy as the error backpropagation algorithm cost function makes the identification procedure using FWNN models more robust to outliers. However, this is only achieved if the gaussian kernel width of correntropy is properly adjusted.

Guiding Monte Carlo tree searches with neural networks in the game of go

A dissertation submitted in fulfillment of the requirements to the degree of Master in Computer Science and Computer Engineering

Tourism demand modelling and forecasting with artificial neural network models: the Mozambique case study

This study is aimed to model and forecast the tourism demand for Mozambique for the period from January 2004 to December 2013 using artificial neural networks models. The number of overnight stays in Hotels was used as representative of the tourism demand. A set of independent variables were experimented in the input of the model, namely: Consumer Price Index, Gross Domestic Product and Exchange Rates, of the outbound tourism markets, South Africa, United State of America, Mozambique, Portugal and the United Kingdom. The best model achieved has 6.5% for Mean Absolute Percentage Error and 0.696 for Pearson correlation coefficient. A model like this with high accuracy of forecast is important for the economic agents to know the future growth of this activity sector, as it is important for stakeholders to provide products, services and infrastructures and for the hotels establishments to adequate its level of capacity to the tourism demand.

A Learning Function for Parameter Reduction in Spiking Neural Networks with Radial Basis Function

Spiking neural networks - networks that encode information in the timing of spikes - are arising as a new approach in the artificial neural networks paradigm, emergent from cognitive science. One of these new models is the pulsed neural network with radial basis function, a network able to store information in the axonal propagation delay of neurons. Learning algorithms have been proposed to this model looking for mapping input pulses into output pulses. Recently, a new method was proposed to encode constant data into a temporal sequence of spikes, stimulating deeper studies in order to establish abilities and frontiers of this new approach. However, a well known problem of this kind of network is the high number of free parameters - more that 15 - to be properly configured or tuned in order to allow network convergence. This work presents for the first time a new learning function for this network training that allow the automatic configuration of one of the key network parameters: the synaptic weight decreasing factor.

Does chemometrics enhance the performance of electroanalysis?

This review explores the question whether chemometrics methods enhance the performance of electroanalytical methods. Electroanalysis has long benefited from the well-established techniques such as potentiometric titrations, polarography and voltammetry, and the more novel ones such as electronic tongues and noses, which have enlarged the scope of applications. The electroanalytical methods have been improved with the application of chemometrics for simultaneous quantitative prediction of analytes or qualitative resolution of complex overlapping responses. Typical methods include partial least squares (PLS), artificial neural networks (ANNs), and multiple curve resolution methods (MCR-ALS, N-PLS and PARAFAC). This review aims to provide the practising analyst with a broad guide to electroanalytical applications supported by chemometrics. In this context, after a general consideration of the use of a number of electroanalytical techniques with the aid of chemometrics methods, several overviews follow with each one focusing on an important field of application such as food, pharmaceuticals, pesticides and the environment. The growth of chemometrics in conjunction with electronic tongue and nose sensors is highlighted, and this is followed by an overview of the use of chemometrics for the resolution of complicated profiles for qualitative identification of analytes, especially with the use of the MCR-ALS methodology. Finally, the performance of electroanalytical methods is compared with that of some spectrophotometric procedures on the basis of figures-of-merit. This showed that electroanalytical methods can perform as well as the spectrophotometric ones. PLS-1 appears to be the method of practical choice if the %relative prediction error of not, vert, similar±10% is acceptable.

Frequency response function based damage identification using principal component analysis and pattern recognition technique

Pattern recognition is a promising approach for the identification of structural damage using measured dynamic data. Much of the research on pattern recognition has employed artificial neural networks (ANNs) and genetic algorithms as systematic ways of matching pattern features. The selection of a damage-sensitive and noise-insensitive pattern feature is important for all structural damage identification methods. Accordingly, a neural networks-based damage detection method using frequency response function (FRF) data is presented in this paper. This method can effectively consider uncertainties of measured data from which training patterns are generated. The proposed method reduces the dimension of the initial FRF data and transforms it into new damage indices and employs an ANN method for the actual damage localization and quantification using recognized damage patterns from the algorithm. In civil engineering applications, the measurement of dynamic response under field conditions always contains noise components from environmental factors. In order to evaluate the performance of the proposed strategy with noise polluted data, noise contaminated measurements are also introduced to the proposed algorithm. ANNs with optimal architecture give minimum training and testing errors and provide precise damage detection results. In order to maximize damage detection results, the optimal architecture of ANN is identified by defining the number of hidden layers and the number of neurons per hidden layer by a trial and error method. In real testing, the number of measurement points and the measurement locations to obtain the structure response are critical for damage detection. Therefore, optimal sensor placement to improve damage identification is also investigated herein. A finite element model of a two storey framed structure is used to train the neural network. It shows accurate performance and gives low error with simulated and noise-contaminated data for single and multiple damage cases. As a result, the proposed method can be used for structural health monitoring and damage detection, particularly for cases where the measurement data is very large. Furthermore, it is suggested that an optimal ANN architecture can detect damage occurrence with good accuracy and can provide damage quantification with reasonable accuracy under varying levels of damage.

Comparison of ANN models for predicting water quality in distribution systems

Deterministic models have been widely used to predict water quality in distribution systems, but their calibration requires extensive and accurate data sets for numerous parameters. In this study, alternative data-driven modeling approaches based on artificial neural networks (ANNs) were used to predict temporal variations of two important characteristics of water quality chlorine residual and biomass concentrations. The authors considered three types of ANN algorithms. Of these, the Levenberg-Marquardt algorithm provided the best results in predicting residual chlorine and biomass with error-free and ``noisy'' data. The ANN models developed here can generate water quality scenarios of piped systems in real time to help utilities determine weak points of low chlorine residual and high biomass concentration and select optimum remedial strategies.

An ANN based approach for voltage stability assessment

Artificial Neural Networks (ANNs) have recently been proposed as an alterative method for salving certain traditional problems in power systems where conventional techniques have not achieved the desired speed, accuracy or efficiency. This paper presents application of ANN where the aim is to achieve fast voltage stability margin assessment of power network in an energy control centre (ECC), with reduced number of appropriate inputs. L-index has been used for assessing voltage stability margin. Investigations are carried out on the influence of information encompassed in input vector and target out put vector, on the learning time and test performance of multi layer perceptron (MLP) based ANN model. LP based algorithm for voltage stability improvement, is used for generating meaningful training patterns in the normal operating range of the system. From the generated set of training patterns, appropriate training patterns are selected based on statistical correlation process, sensitivity matrix approach, contingency ranking approach and concentric relaxation method. Simulation results on a 24 bus EHV system, 30 bus modified IEEE system, and a 82 bus Indian power network are presented for illustration purposes.

An approach to multi-temporal MODIS image analysis using image classification and segmentation

This paper discusses an approach for river mapping and flood evaluation based on multi-temporal time series analysis of satellite images utilizing pixel spectral information for image classification and region-based segmentation for extracting water-covered regions. Analysis of MODIS satellite images is applied in three stages: before flood, during flood and after flood. Water regions are extracted from the MODIS images using image classification (based on spectral information) and image segmentation (based on spatial information). Multi-temporal MODIS images from ``normal'' (non-flood) and flood time-periods are processed in two steps. In the first step, image classifiers such as Support Vector Machines (SVMs) and Artificial Neural Networks (ANNs) separate the image pixels into water and non-water groups based on their spectral features. The classified image is then segmented using spatial features of the water pixels to remove the misclassified water. From the results obtained, we evaluate the performance of the method and conclude that the use of image classification (SVM and ANN) and region-based image segmentation is an accurate and reliable approach for the extraction of water-covered regions. (c) 2012 COSPAR. Published by Elsevier Ltd. All rights reserved.

A study on the usefulness of Support Vector Machines for the realtime computational simulation of soft biological organs

Realistic and realtime computational simulation of soft biological organs (e.g., liver, kidney) is necessary when one tries to build a quality surgical simulator that can simulate surgical procedures involving these organs. Since the realistic simulation of these soft biological organs should account for both nonlinear material behavior and large deformation, achieving realistic simulations in realtime using continuum mechanics based numerical techniques necessitates the use of a supercomputer or a high end computer cluster which are costly. Hence there is a need to employ soft computing techniques like Support Vector Machines (SVMs) which can do function approximation, and hence could achieve physically realistic simulations in realtime by making use of just a desktop computer. Present work tries to simulate a pig liver in realtime. Liver is assumed to be homogeneous, isotropic, and hyperelastic. Hyperelastic material constants are taken from the literature. An SVM is employed to achieve realistic simulations in realtime, using just a desktop computer. The code for the SVM is obtained from [1]. The SVM is trained using the dataset generated by performing hyperelastic analyses on the liver geometry, using the commercial finite element software package ANSYS. The methodology followed in the present work closely follows the one followed in [2] except that [2] uses Artificial Neural Networks (ANNs) while the present work uses SVMs to achieve realistic simulations in realtime. Results indicate the speed and accuracy that is obtained by employing the SVM for the targeted realistic and realtime simulation of the liver.

Use of Data Driven Techniques for Short Lead Time Streamflow Forecasting in Mahanadi basin

Streamflow forecasts at daily time scale are necessary for effective management of water resources systems. Typical applications include flood control, water quality management, water supply to multiple stakeholders, hydropower and irrigation systems. Conventionally physically based conceptual models and data-driven models are used for forecasting streamflows. Conceptual models require detailed understanding of physical processes governing the system being modeled. Major constraints in developing effective conceptual models are sparse hydrometric gauge network and short historical records that limit our understanding of physical processes. On the other hand, data-driven models rely solely on previous hydrological and meteorological data without directly taking into account the underlying physical processes. Among various data driven models Auto Regressive Integrated Moving Average (ARIMA), Artificial Neural Networks (ANNs) are most widely used techniques. The present study assesses performance of ARIMA and ANNs methods in arriving at one-to seven-day ahead forecast of daily streamflows at Basantpur streamgauge site that is situated at upstream of Hirakud Dam in Mahanadi river basin, India. The ANNs considered include Feed-Forward back propagation Neural Network (FFNN) and Radial Basis Neural Network (RBNN). Daily streamflow forecasts at Basantpur site find use in management of water from Hirakud reservoir. (C) 2015 The Authors. Published by Elsevier B.V.

Virtual Sensors for On-line Wheel Wear and Part Roughness Measurement in the Grinding Process

Grinding is an advanced machining process for the manufacturing of valuable complex and accurate parts for high added value sectors such as aerospace, wind generation, etc. Due to the extremely severe conditions inside grinding machines, critical process variables such as part surface finish or grinding wheel wear cannot be easily and cheaply measured on-line. In this paper a virtual sensor for on-line monitoring of those variables is presented. The sensor is based on the modelling ability of Artificial Neural Networks (ANNs) for stochastic and non-linear processes such as grinding; the selected architecture is the Layer-Recurrent neural network. The sensor makes use of the relation between the variables to be measured and power consumption in the wheel spindle, which can be easily measured. A sensor calibration methodology is presented, and the levels of error that can be expected are discussed. Validation of the new sensor is carried out by comparing the sensor's results with actual measurements carried out in an industrial grinding machine. Results show excellent estimation performance for both wheel wear and surface roughness. In the case of wheel wear, the absolute error is within the range of microns (average value 32 mu m). In the case of surface finish, the absolute error is well below R-a 1 mu m (average value 0.32 mu m). The present approach can be easily generalized to other grinding operations.

Inteligência computacional aplicada na geração de respostas impulsivas bi-auriculares e em aurilização de salas

Neste trabalho é apresentada uma nova abordagem para obter as respostas impulsivas biauriculares (BIRs) para um sistema de aurilização utilizando um conjunto de redes neurais artificiais (RNAs). O método proposto é capaz de reconstruir as respostas impulsivas associadas à cabeça humana (HRIRs) por meio de modificação espectral e de interpolação espacial. A fim de cobrir todo o espaço auditivo de recepção, sem aumentar a complexidade da arquitetura da rede, uma estrutura com múltiplas RNAs (conjunto) foi adotada, onde cada rede opera uma região específica do espaço (gomo). Os três principais fatores que influenciam na precisão do modelo arquitetura da rede, ângulos de abertura da área de recepção e atrasos das HRIRs são investigados e uma configuração ideal é apresentada. O erro de modelagem no domínio da frequência é investigado considerando a natureza logarítmica da audição humana. Mais ainda, são propostos novos parâmetros para avaliação do erro, definidos em analogia com alguns dos bem conhecidos parâmetros de qualidade acústica de salas. Através da metodologia proposta obteve-se um ganho computacional, em redução do tempo de processamento, de aproximadamente 62% em relação ao método tradicional de processamento de sinais utilizado para aurilização. A aplicabilidade do novo método em sistemas de aurilização é reforçada mediante uma análise comparativa dos resultados, que incluem a geração das BIRs e o cálculo dos parâmetros acústicos biauriculares (IACF e IACC), os quais mostram erros de magnitudes reduzidas.