The paper describes the development and application of several techniques for knowledge extraction from trained ANN models, such as the identification of redundant inputs and hidden neurons, deriving of causal relationships between inputs and outputs, and analysis of the hidden neuron behavior in classification ANN. Example of the application of these techniques is given of the faulty LED displ...

In this paper, we present a new approach towards user-customized password speaker verification combining the advantages of hybrid HMM/ANN systems, usingArtificial Neural Networks (ANN) to estimate emission probabilities of Hidden Markov Models , and Gaussian Mixture Models. In the approach presented here, we indeed exploit the properties of hybrid HMM/ANN systems, usually resulting in high phon...

The results of two alternative models to predict segmental durations in speech synthesis, both based on Artificial Neural Networks (ANNs) are discussed. The ANN model consists in just one ANN trained to predict the segmental durations for all phonemes. The phoneme dedicated ANN model consists in a set of ANNs, each one dedicated to predict the segmental duration of a specific phoneme. Both mode...

The present research work is about to disaster mitigation using the applications of ANN. The ANN is used in the number of diverse fields due to its ability to model non linear patterns and self adjusting (learning) nature to produce consistent output when trained using supervised learning. This study utilizes Backpropagation Neural Network to train ANN models to mitigation of disaster through f...

While Artificial Neural Networks (ANNs) are highly expressive models, they are hard to train from limited data. Formalizing a connection between Random Forests (RFs) and ANNs allows exploiting the former to initialize the latter. Further parameter optimization within the ANN framework yields models that are intermediate between RF and ANN, and achieve performance better than RF and ANN on the m...

As a versatile investment tool in energy markets for speculators and hedgers, the Goldman Sachs Commodity Index (GSCI) futures are quite well known. Therefore, this paper proposes a hybrid model incorporating ARCH family models and ANN model to forecast GSCI futures price. Empirical results show that the hybrid ARCH(1)-M-ANN model is superior to ARIMA, ARCH(1),GARCH(1,1), EGARCH(1,1) and ARIMA-...

The back-propagation (BP) artificial neural network (ANN) is applied to forecast the variation of the quality of groundwater in the blackfoot disease area in Taiwan. Three types of BP ANN models were established to evaluate their learning performance. Model A included five concentration parameters as input variables for seawater intrusion and three concentration parameters as input variables fo...

BACKGROUND The present study aimed to develop an artificial neural network (ANN) based prediction model for cardiovascular autonomic (CA) dysfunction in the general population. METHODS We analyzed a previous dataset based on a population sample consisted of 2,092 individuals aged 30-80 years. The prediction models were derived from an exploratory set using ANN analysis. Performances of these ...

This work develops Artificial Neural Networks (ANN) models applied to predict the consumption forecasting considering climatic factors. It is intended to verify the influence of climatic factors on the electricity consumption forecasting through the ANN. The case study is applied in the Campinas city, Brazil. This work used Perceptron and Backpropagation ANN models. The specific goal is compari...

BACKGROUND Since most published articles comparing the performance of artificial neural network (ANN) models and logistic regression (LR) models for predicting hepatocellular carcinoma (HCC) outcomes used only a single dataset, the essential issue of internal validity (reproducibility) of the models has not been addressed. The study purposes to validate the use of ANN model for predicting in-ho...