Multilogistic regression by means of evolutionary product-unit neural networks

Hervás-Martı́nez, César; Martínez-Estudillo, Francisco José; Carbonero-Ruz, Mariano

doi:10.1016/j.neunet.2007.12.052

Cited by 43 publications

(21 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we describe the main characteristics of the binary LR approach and the hybrid models considered. Note that a complete description of the EPUNNs and ERBFNNs is not carried out due to space reasons (the reader can check the references [11,12,13] for a complete description of the models and the EA used to optimize these models).…”

Section: Description Of the Hybrid Methodologies Proposedmentioning

confidence: 99%

“…. , w jk ), w ji ∈ R. The coefficients W are given by the EA, they not being adjusted by the ML method (the reader can check the references [11,12,13] for a complete description of this EA). The ML method only optimizes the linear part of the model, i.e.…”

Section: Logistic Regression Using Initial Covariates and Product Unimentioning

confidence: 99%

“…Logistic Regression (LR) has become a widely used and accepted method of analysis of binary or multi-class outcome variables as it is flexible and it can predict the probability for the state of a dichotomous variable (in our case, the probability of crisis) based on the predictor variables (macroeconomic variables in this case). The hybridization of LR and PUNNs or RBFNNs is done following the model of classifier construction given in [11], where Evolutionary Algorithms (EAs) are used to obtain the best structure of the classifiers. In this paper we show that the hybrid models involving LR, PUNNs and RBFNNs outperforms several other existing classification techniques in the problem of banking crises prediction, and they are therefore a very interesting tool to take into account in this field.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Generalized Logistic Regression Models Using Neural Network Basis Functions Applied to the Detection of Banking Crises

Gutiérrez

Salcedo‐Sanz

Vargas

et al. 2010

Trends in Applied Intelligent Systems

Self Cite

View full text Add to dashboard Cite

Abstract. The financial system plays a crucial role in economic development. Financial crises are recurrent phenomena in modern financial systems. The literature offers several definitions of financial instability, but it is well known that a financial crisis with a banking crisis is the most common example of financial instability. In this paper we introduce a novel model for detection and prediction of crises, based on the hybridization of a standard logistic regression with Product Unit (PU) neural networks and Radial Basis Function (RBF) networks. These hybrid approaches are described in the paper, and applied to the detection and prediction of banking crises by using a large database of countries in the period 1981 to 1999. The proposed techniques are shown to perform better than other existing statistical and artificial intelligence methods for this problem.

show abstract

Section: Description Of the Hybrid Methodologies Proposedmentioning

confidence: 99%

Section: Logistic Regression Using Initial Covariates and Product Unimentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Generalized Logistic Regression Models Using Neural Network Basis Functions Applied to the Detection of Banking Crises

Gutiérrez

Salcedo‐Sanz

Vargas

et al. 2010

Trends in Applied Intelligent Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…1, each training set P , is divided into n bags (3), so that the weights of the individual which is being evaluated are applied to n − 1 bags (5), and the remaining is used as initial test (6 et seq.). The transformation of each label is applied to each pixel of the test bag (6-8) and then, the nearest pixel from P is calculated (9). Once the point has been tested, it becomes part of P reinforcing the training (10).…”

Section: Fitness Functionmentioning

confidence: 99%

“…Additionally, there are basically three main areas of weighting application in supervised machine learning: support vector machines optimization, artificial neural networks (training and topology) and feature weighting. Thus, SVM kernel [8] or artificial neural networks [9] parameters can be optimized by means of genetic algorithms or genetic programming with good results. In this context, evolutionary algorithms are usually employed to find a set of weights for the feature space, allowing greater accuracy in the classification process [10].…”

Section: Introductionmentioning

confidence: 99%

Label Dependent Evolutionary Feature Weighting for Remote Sensing Data

Mateos-García

García-Gutiérrez

Santos

2010

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Nearest neighbour (NN) is a very common classifier used to develop important remote sensing products like land use and land cover (LULC) maps. Evolutive computation has often been used to obtain feature weighting in order to improve the results of the NN. In this paper, a new algorithm based on evolutionary computation which has been called Label Dependent Feature Weighting (LDFW) is proposed. The LDFW method transforms the feature space assigning different weights to every feature depending on each class. This multilevel feature weighting algorithm is tested on remote sensing data from fusion of sensors (LIDAR and orthophotography). The results show an improvement on the NN and resemble the results obtained with a neural network which is the best classifier for the study area.

show abstract

Neural network‐based approaches for building high dimensional and quantum dynamics‐friendly potential energy surfaces

Manzhos

Dawes

Carrington

2014

Int J of Quantum Chemistry

196

148

View full text Add to dashboard Cite

Development and applications of neural network (NN)-based approaches for representing potential energy surfaces (PES) of bound and reactive molecular systems are reviewed. Specifically, it is shown that when the density of ab initio points is low, NNs-based potentials with multibody or multimode structure are advantageous for representing high-dimensional PESs. Importantly, with an appropriate choice of the neuron activation function, PESs in the sum-of-products form are naturally obtained, thus addressing a bottleneck problem in quantum dynamics. The use of NN committees is also analyzed and it is shown that while they are able to reduce the fitting error, the reduction is limited by the nonrandom nature of the fitting error. The approaches described here are expected to be directly applicable in other areas of science and engineering where a functional form needs to be constructed in an unbiased way from sparse data.

show abstract

Multilogistic regression by means of evolutionary product-unit neural networks

Cited by 43 publications

References 40 publications

Generalized Logistic Regression Models Using Neural Network Basis Functions Applied to the Detection of Banking Crises

Generalized Logistic Regression Models Using Neural Network Basis Functions Applied to the Detection of Banking Crises

Label Dependent Evolutionary Feature Weighting for Remote Sensing Data

Neural network‐based approaches for building high dimensional and quantum dynamics‐friendly potential energy surfaces

Contact Info

Product

Resources

About