Search

Abstract: Estimating the prediction error of classifiers induced by supervised learning algorithms is important not only to predict its future error, but also to choose a classifier from a given set (model selection). If the goal is to estimate the prediction error of a particular classifier, the desired estimator should have low bias and low variance. However, if the goal is the model selection, in order to make fair comparisons the chosen estimator should have low variance assuming that the bias term is independent from the considered classifier. This paper follows the analysis proposed in [1] about the statistical properties of k-fold cross-validation estimators and extends it to the most popular error estimators: resubstitution, holdout, repeated holdout, simple bootstrap and 0.632 bootstrap estimators, without and with stratification. We present a general framework to analyze the decomposition of the variance of different error estimators considering the nature of the variance (irreducible/reducible variance) and the different sources of sensitivity (internal/external sensitivity). An extensive empirical study has been performed for the previously mentioned estimators with naive Bayes and C4.5 classifiers over training sets obtained from assorted probability distributions. The empirical analysis consists of decomposing the variances following the proposed framework and checking the independence assumption between the bias and the considered classifier. Based on the obtained results, we propose the most appropriate error estimations for model selection under different experimental conditions. [Copyright &y& Elsevier]

Copyright of Revista Española de Cardiología (18855857) is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Growth, biomass yield, fiber content and lodging resistance were studied, during a six month growth period, for eight varieties of Cenchrus purpureus , intended as energy crop, in Veracruz, Mexico. Then, only yield at day 182 was assessed for two additional years. The varieties were: CT115 (CT), African Cane (AC), Taiwan (TAI), King Grass (KG), Vruckwona (VRU), Roxo (RX), OM22 (OM) and Cameroon (CAM). Local weather is warm and sub-humid, historical data for monthly average temperature and annual rainfall were 25.8 °C and 1142 mm, respectively. Height, diameter and light interception were measured monthly from day 65–185. At day 185, biomass yield and tiller density were measured. Number of lying tillers was counted to estimate lodging resistance. Cellulose and hemicellulose content were estimated in leaf and stem. No differences were found for dry matter yield or stem yield at day 185 in the first year. Regarding the next two years, TAI yielded above CT, OM or ROX. Average dry matter yield was higher in the second year than in the establishment cycle (38.6 vs 21.1 Mg ha −1 ), but decreased in the third year (32.2 Mg ha −1 ). In both stem and whole plant, AC and KG showed higher hemicellulose content than RX, OM or CT; while AC and VRU had higher cellulose than RX in stem, or than CT in the whole plant. Furthermore, varieties AC, KG, VRU and TAI were resistant to lodging and had a higher fiber content, so they are recommended as energetic crops. [ABSTRACT FROM AUTHOR]

Abstract: A multi-species approach to fisheries management requires taking into account the interactions between species in order to improve recruitment forecasting of the fish species. Recent advances in Bayesian networks direct the learning of models with several interrelated variables to be forecasted simultaneously. These models are known as multi-dimensional Bayesian network classifiers (MDBNs). Pre-processing steps are critical for the posterior learning of the model in these kinds of domains. Therefore, in the present study, a set of ‘state-of-the-art’ uni-dimensional pre-processing methods, within the categories of missing data imputation, feature discretization and feature subset selection, are adapted to be used with MDBNs. A framework that includes the proposed multi-dimensional supervised pre-processing methods, coupled with a MDBN classifier, is tested with synthetic datasets and the real domain of fish recruitment forecasting. The correctly forecasting of three fish species (anchovy, sardine and hake) simultaneously is doubled (from 17.3% to 29.5%) using the multi-dimensional approach in comparison to mono-species models. The probability assessments also show high improvement reducing the average error (estimated by means of Brier score) from 0.35 to 0.27. Finally, these differences are superior to the forecasting of species by pairs. [Copyright &y& Elsevier]