Bertha Guijarro-Berdiñas

A global optimum approach for one-layer neural networks

The article presents a method for learning the weights in one-layer feed-forward neural networks minimizing either the sum of squared errors or the maximum absolute error, measured in the input scale. This leads to the existence of a global optimum that can be easily obtained solving linear systems of equations or linear programming problems, using much less computational power than the one associated with the standard methods. Another version of the method allows computing a large set of estimates for the weights, providing robust, mean or median, estimates for them, and the associated standard errors, which give a good measure for the quality of the fit. Later, the standard one-layer neural network algorithms are improved by learning the neural functions instead of assuming them known. A set of examples of applications is used to illustrate the methods. Finally, a comparison with other high-performance learning algorithms shows that the proposed methods are at least 10 times faster than the fastest standard algorithm used in the comparison.

An intelligent system for forest fire risk prediction and fire fighting management in Galicia

Abstract Over the last two decades in southern Europe, more than 10 million hectares of forest have been damaged by fire. Due to the costs and complications of fire-fighting a number of technical developments in the field have been appeared in recent years. This paper describes a system developed for the region of Galicia in NW Spain, one of the regions of Europe most affected by fires. This system fulfills three main aims: it acts as a preventive tool by predicting forest fire risks, it backs up the forest fire monitoring and extinction phase, and it assists in planning the recuperation of the burned areas. The forest fire prediction model is based on a neural network whose output is classified into four symbolic risk categories, obtaining an accuracy of 0.789. The other two main tasks are carried out by a knowledge-based system developed following the CommonKADS methodology. Currently we are working on the trail of the system in a controlled real environment. This will provide results on real behaviour that can be used to fine-tune the system to the point where it is considered suitable for installation in a real application environment.

A new method for sleep apnea classification using wavelets and feedforward neural networks

OBJECTIVES This paper presents a novel approach for sleep apnea classification. The goal is to classify each apnea in one of three basic types: obstructive, central and mixed. MATERIALS AND METHODS Three different supervised learning methods using a neural network were tested. The inputs of the neural network are the first level-5-detail coefficients obtained from a discrete wavelet transformation of the samples (previously detected as apnea) in the thoracic effort signal. In order to train and test the systems, 120 events from six different patients were used. The true error rate was estimated using a 10-fold cross validation. The results presented in this work were averaged over 100 different simulations and a multiple comparison procedure was used for model selection. RESULTS The method finally selected is based on a feedforward neural network trained using the Bayesian framework and a cross-entropy error function. The mean classification accuracy, obtained over the test set was 83.78+/-1.90%. CONCLUSION The proposed classifier surpasses, up to the authors knowledge, other previous results. Finally, a scheme to maintain and improve this system during its clinical use is also proposed.

Intelligent analysis and pattern recognition in cardiotocographic signals using a tightly coupled hybrid system

In obstetrics, cardiotocograph (CTG) and non-stress test readings are indispensable to antenatal monitoring and assessment. Difficulties in the interpretation of CTG records require methods for computer-assisted analysis. This article describes CAFE (Computer Aided Foetal Evaluator), an intelligent tightly coupled hybrid system developed to overcome the difficulties inherent in CTG analysis. It integrates algorithms (implemented via conventional programming techniques) with Artificial Intelligence (AI) paradigms (rule-based systems and artificial neural networks), in order to automate and perform all the phases involved in real time antenatal monitoring, from the analysis and interpretation of CTG signals to diagnosis. Its architecture, components and functional character will be described in detail. The validation of CAFE over 3450 minutes of signal time corresponding to 53 different patients in a real environment is discussed, and its performance with respect to a group of experts is evaluated. Most of the results obtained reflect acceptable levels of performanceequivalent to expert performanceand thus confirm the suitability of AI techniques to applications in this field.

A survey of methods for distributed machine learning

Traditionally, a bottleneck preventing the development of more intelligent systems was the limited amount of data available. Nowadays, the total amount of information is almost incalculable and automatic data analyzers are even more needed. However, the limiting factor is the inability of learning algorithms to use all the data to learn within a reasonable time. In order to handle this problem, a new field in machine learning has emerged: large-scale learning. In this context, distributed learning seems to be a promising line of research since allocating the learning process among several workstations is a natural way of scaling up learning algorithms. Moreover, it allows to deal with data sets that are naturally distributed, a frequent situation in many real applications. This study provides some background regarding the advantages of distributed environments as well as an overview of distributed learning for dealing with “very large” data sets.

A new convex objective function for the supervised learning of single-layer neural networks

This paper proposes a novel supervised learning method for single-layer feedforward neural networks. This approach uses an alternative objective function to that based on the MSE, which measures the errors before the neurons nonlinear activation functions instead of after them. In this case, the solution can be easily obtained solving systems of linear equations, i.e., requiring much less computational power than the one associated with the regular methods. A theoretical study is included to proof the approximated equivalence between the global optimum of the objective function based on the regular MSE criterion and the one of the proposed alternative MSE function. Furthermore, it is shown that the presented method has the capability of allowing incremental and distributed learning. An exhaustive experimental study is also presented to verify the soundness and efficiency of the method. This study contains 10 classification and 16 regression problems. In addition, a comparison with other high performance learning algorithms shows that the proposed method exhibits, in average, the highest performance and low-demanding computational requirements.

The PATRICIA project: a semantic-based methodology for intelligent monitoring in the ICU

The authors describe PATRICIA, an intelligent monitoring system designed to advise clinicians in the management of patients dependent on mechanical ventilation. PATRICIA incorporates a patient-oriented symbolic approach for representing knowledge and a symbolic-oriented temporal approach for the intelligent control of the monitoring process. The results and methodology used in the validation of PATRICIA are presented. Preliminary results show that PATRICIA could be a useful tool for the management of patients receiving mechanical ventilatory support.<<ETX>>

A linear learning method for multilayer perceptrons using least-squares

Training multilayer neural networks is typically carried out using gradient descent techniques. Ever since the brilliant backpropagation (BP), the first gradient-based algorithm proposed by Rumelhart et al., novel training algorithms have appeared to become better several facets of the learning process for feed-forward neural networks. Learning speed is one of these. In this paper, a learning algorithm that applies linear-least-squares is presented. We offer the theoretical basis for the method and its performance is illustrated by its application to several examples in which it is compared with other learning algorithms and well known data sets. Results show that the new algorithm upgrades the learning speed of several backpropagation algorithms, while preserving good optimization accuracy. Due to its performance and low computational cost it is an interesting alternative, even for second order methods, particularly when dealing large networks and training sets.

Empirical evaluation of a hybrid intelligent monitoring system using different measures of effectiveness

The validation of a software product is a fundamental part of its development, and focuses on an analysis of whether the software correctly resolves the problems it was designed to tackle. Traditional approaches to validation are based on a comparison of results with what is called a gold standard. Nevertheless, in certain domains, it is not always easy or even possible to establish such a standard. This is the case of intelligent systems that endeavour to simulate or emulate a model of expert behaviour. This article describes the validation of the intelligent system computer-aided foetal evaluator (CAFE), developed for intelligent monitoring of the antenatal condition based on data from the non-stress test (NST), and how this validation was accomplished through a methodology designed to resolve the problem of the validation of intelligent systems. System performance was compared to that of three obstetricians using 3450 min of cardiotocographic (CTG) records corresponding to 53 different patients. From these records different parameters were extracted and interpreted, and thus, the validation was carried out on a parameter-by-parameter basis using measurement techniques such as percentage agreement, the Kappa statistic or cluster analysis. Results showed that the systems agreement with the experts is, in general, similar to agreement between the experts themselves which, in turn, permits our system to be considered at least as skillful as our experts. Throughout our article, the results obtained are commented on with a view to demonstrating how the utilisation of different measures of the level of agreement existing between system and experts can assist not only in assessing the aptness of a system, but also in highlighting its weaknesses. This kind of assessment means that the system can be fine-tuned repeatedly to the point where the expected results are obtained.

Adaptive pattern recognition in the analysis of cardiotocographic records

The recognition of accelerative and decelerative patterns in the fetal heart rate (FHR) is one of the tasks carried out manually by obstetricians when they analyze cardiotocograms for information respecting the fetal state. An approach based on artificial neural networks formed by a multilayer perceptron (MLP) is developed. However, since the system utilizes the FHR signal as direct input, an anterior stage must be incorporated that applies a principal component analysis (PCA) so as to make the system independent of the signal baseline. Furthermore, the introduction of multiresolution into the PCA has resolved other problems that were detected in the application of the system. Presented in this paper are the results of validation of these systems designated the PCA-MLP and multiresolutlon principal component analysis (MR-PCA) systems against three clinical experts.