Abstract not found

Abstract. This paper proposes a methodology named OP-ELM, based on a recent development –the Extreme Learning Machine – decreasing drastically the training speed of networks. Variable selection is beforehand performed on the original dataset for proper results by OP-ELM: the network is first created using Extreme Learning Process, selection of the most relevant nodes is performed using Least Angle Regression (LARS) ranking of the nodes and a Leave-One-Out estimation of the performances. Results are globally equivalent to LSSVM ones with reduced computational time. 1

Abstract. This paper presents a methodology for functional data analysis. It consists in extracting a large number of features with maximal content of information and then selecting the appropriate ones through a Mutual Information criterion; next, this reduced set of features is used to build a classifier. The methodology is applied to an industrial problem: the classification of the dynamic properties of elastomeric material characterized by rigidity and hysteresis curves. 1

Abstract. In spectrometric problems, objects are characterized by highresolution spectra that correspond to hundreds to thousands of variables. In this context, even fast variable selection methods lead to high computational load. However, spectra are generally smooth and can therefore be accurately approximated by splines. In this paper, we propose to use a B-spline expansion as a pre-processing step before variable selection, in which original variables are replaced by coefficients of the B-spline expansions. Using a simple leave-one-out procedure, the optimal number of B-spline coefficients can be found efficiently. As there is generally an order of magnitude less coefficients than original spectral variables, selecting optimal coefficients is faster than selecting variables. Moreover, a B-spline coefficient depends only on a limited range of original variables: this preserves interpretability of the selected variables. We demonstrate the interest of the proposed method on real-world data. 1

Abstract. Input variable selection is a key preprocess step in any I/O modelling problem. Normally, better generalization performance is obtained when unneeded parameters coming from irrelevant or redundant variables are eliminated. Information theory provides a robust theoretical framework for performing input variable selection thanks to the concept of mutual information. Nevertheless, for continuous variables, it is usually a more difficult task to determine the mutual information between the input variables and the output variable than for classification problems. This paper presents a modified approach for variable selection for continuous variables adapted from a previous approach for classification problems, making use of a mutual information estimator based on the k-nearest neighbors. 1

Abstract. Spectral data often have a large number of highly-correlated features, making feature selection both necessary and uneasy. A methodology combining hierarchical constrained clustering of spectral variables and selection of clusters by mutual information is proposed. The clustering allows reducing the number of features to be selected by grouping similar and consecutive spectral variables together, allowing an easy interpretation. The approach is applied to two datasets related to spectroscopy data from the food industry. 1

Selecting relevant features in mass spectra analysis is important both for classification and search for causality. In this paper, it is shown how using mutual information can help answering to both objectives, in a model-free nonlinear way. A combination of ranking and forward selection makes it possible to select several feature groups that may lead to similar classification performances, but that may lead to different results when evaluated from an interpretability perspective. 1.

Abstract. We study filter–based feature selection methods for classification of biomedical images. For feature selection, we use two filters — a relevance filter which measures usefulness of individual features for target prediction, and a redundancy filter, which measures similarity between features. As selection method that combines relevance and redundancy we try out a Hopfield network. We experimentally compare selection methods, running unitary redundancy and relevance filters, against a greedy algorithm with redundancy thresholds [9], the min-redundancy max-relevance integration [8,23,36], and our Hopfield network selection. We conclude that on the whole, Hopfield selection was one of the most successful methods, outperforming min-redundancy max-relevance when more features are selected. Key words: feature selection, image features, pattern classification 1

Abstract. In this paper, we propose a new methodology to build latent variables that are optimal if a nonlinear model is used afterward. This method is based on Nonparametric Noise Estimation (NNE). NNE is providing an estimate of the variance of the noise between input and output variables. The linear projection that builds latent variables is optimized in order to minimize the NNE. We successfully tested the proposed methodology on a referenced spectral dataset from food industry (Tecator). 1

Abstract not found