In this paper, we describe a method for learning shape descriptions of objects in x-ray images. The descriptions are induced from shape examples using the AQ15c inductive learning system. The method has been experimentally compared to k-nearest neighbor, a statistical pattern recognition technique, the C4.5 decision tree learning program, and a multilayer feed-forward neural network. Experiment...

The removal of acid gases from gas streams by using suitable solvent like alkanolamine, commonly referred to as gas sweetening, is a technology that has been in use industrially for over half a century. In this work artificial neural network (ANN) has been used to predict the equilibrium solubility of CO2 over the alkanolamine solvents N-methyldiethanolamine (MDEA) and 2-amino-2-methyl-1-propan...

We describe a new approach to multiple class pattern classification problems with noise and high dimensional feature space. The approach uses a random matrix X which has a specified distribution with mean M and covariance matrix ri j(Σs +Σ ) between any two columns of X . When Σ is known, the maximum likelihood estimators of the expectation M, correlation Γ, and covariance Σs can be obtained. T...

Abstract— To expand ontology meanings, an effective ontology mapping approach is needed to map related or similar knowledge from heterogeneous sources together. Especially, the mapping approach also can be applied to support image recognition in order to enhance its retrieval information. In this paper, we propose the ontology mapping with back propagation method to learn image objects, and lin...

In this paper, we purpose a diagnostic procedure to identify the macular disease from pattern electroretionography (PERG) signals using artificial neural networks (ANN) methods. Multilayer feed forward ANN trained with a Levenberg Marquart backpropagation algorithm was implemented. The designed classification structure has about 96% sensitivity, 100% specifity and correct classification is calc...

Signature verification is most commonly used as an authorization tool from the beginning till now. Many people uses bank cheques for most of their transactions. Although banks are computerized, but still verification process of signature in cheques is done manually which consumes time and even misleads sometimes. Signatures verification process can be done online or off-line depending upon the ...

The performance of recently developed neural network structure, general regression neural network (GRNN), is examined on the medical data. Pima Indian Dabetes (PID) data set is chosen to study on that had been examined by more complex neural network structures in the past. The results of early studies and of the GRNN structure presented in this paper is compared. Close classification accuracy t...

In a previous work, we showed that artificial neural networks (ANNs) could learn the criteria for comparing fuzzy numbers of a real decision maker. A multilayer feed-forward ANN and the backpropagation algorithm, and trapezoidal fuzzy numbers were considered. The criteria of three people were learnt with an ANN. The trained ANN is considered as a personal method of the decision-maker to compare...

The automatic prosodic annotation of large speech corpora gains increasing consideration since appropriate databases for the training of prosodic models in speech synthesis and recognition are needed. On linguistic level, correct phrase and accent marking are essential processing steps. The authors developed a neural network based method for signal-based phrase break prediction and tested this ...

A handwritten character recognition system using multilayer Feed forward neural network is proposed in this paper. The character data set suitable for recognizing postal addresses contains 38 elements which include 26 alphabets, 10 numerals and 2 symbols. Fifteen different handwritten data sets were used for training the neural network for classification and recognition of the characters. Three...