Risk bounds for CART classifiers under a margin condition

Non-asymptotic risk bounds for Classification And Regression Trees (CART) classifiers are obtained in the binary supervised classification framework under a margin assumption on the joint distribution of the covariates and the labels. These risk bounds are derived conditionally on the construction of the maximal binary tree and allow to prove that the linear penalty used in the CART pruning algorithm is valid under the margin condition.It is also shown that, conditionally on the construction of the maximal tree, the final selection by test sample does not alter dramatically the estimation accuracy of the Bayes classifier.     

How to improve the hygienic quality of forages for horse feeding

BACKGROUND: Improving the hygienic quality of forages for horse nutrition seems to be a reasonable target for decreasing the prevalence of pulmonary diseases. The aim of the experiment was to study the effects of different agricultural practices on the main aero‐allergens contained in forages, including breathable dust, fungi, mycotoxins and pollens. RESULTS: Results showed that the late harvest of hay, a second crop or a haylage production provides a good alternative to increase hygienic quality by reducing fungi contamination and breathable dust content. Barn drying of hay, while having no effect on breathable dust, similarly reduced fungi contamination. In contrast, when hay was harvested at a lower dry mass content (750 g DM kg^?1 versus 850 g DM kg^?1), both breathable dust and fungi contaminations were increased, which could at least be reversed by adding propionic acid just before baling. Zearalenone was detected in different hays, and even in one case, in breathable dust. CONCLUSION: Overall, our data suggest that different approaches can be used to increase forage hygienic quality for horse feeding and thus reduce their exposure to factors involved in equine pulmonary disease. Copyright © 2011 Society of Chemical Industry     

Depth profiling of high-energy hydrogen-implanted 6H-SiC

The results of implanting silicon carbide with a 1-MeV proton beam at a dose of 1 x 10(17) cm(-2) are presented. Using high-resolution confocal Raman spectroscopy, we analyzed the depth profile of the implantation damage before and after thermal annealing. When it is applied to a high-refractive-index medium, such as SiC, this technique requires careful manipulation to ensure the correct interpretation of results. To this end we discuss a simple ray-tracking model that includes the effects of additional spherical aberration and of the Gaussian intensity profile of the excitation beam. In addition, infrared reflectance measurements show evidence of a well-defined step in the refractive-index profile at the expected implantation depth.