A novel information theory method for filter feature selection

In this paper, we propose a novel filter for feature selection. Such filter relies on the estimation of the mutual information between features and classes. We bypass the estimation of the probability density function with the aid of the entropic-graphs approximation of Rényi entropy, and the subsequent approximation of the Shannon one. The complexity of such bypassing process does not depend on the number of dimensions but on the number of patterns/samples, and thus the curse of dimensionality is circumvented. We show that it is then possible to outperform a greedy algorithm based on the maximal relevance and minimal redundancy criterion. We successfully test our method both in the contexts of image classification and microarray data classification.

Risco, corpo e socialidade no voo livre

Universidade Estadual de Campinas . Faculdade de Educação Física

THE RELEVANCE OF SUBSIDIARY INITIATIVES FOR BRAZILIAN MULTINATIONALS

The put-pose of this paper is to analyze relationship patterns between headquarters and subsidiaries of Brazilian Multinationals Enterprises (BrMNEs). The key construct for that investigation is Subsidiary Initiative, which comprises Subsidiary Entrepreneurial Orientation, Autonomy, Integration, Local Competitive Context and Business Network. A survey was carried out in a sample of 65 subsidiaries of 29 BrMNEs. The main outcome is that subsidiaries are highly integrated and receive Entrepreneurial Orientation from Headquarters (HQs), but Initiative is limited. Actually, the main determinants of subsidiary's initiatives are Local Context and Business Networking in the host country. This apparent paradox may be explained by what we call 'rebellious subsidiaries', which take initiatives based on their business environment and connections, regardless of their HQs' directions or delegation of autonomy.

Characterization of subgraph relationships and distribution in complex networks

A network can be analyzed at different topological scales, ranging from single nodes to motifs, communities, up to the complete structure. We propose a novel approach which extends from single nodes to the whole network level by considering non-overlapping subgraphs (i.e. connected components) and their interrelationships and distribution through the network. Though such subgraphs can be completely general, our methodology focuses on the cases in which the nodes of these subgraphs share some special feature, such as being critical for the proper operation of the network. The methodology of subgraph characterization involves two main aspects: (i) the generation of histograms of subgraph sizes and distances between subgraphs and (ii) a merging algorithm, developed to assess the relevance of nodes outside subgraphs by progressively merging subgraphs until the whole network is covered. The latter procedure complements the histograms by taking into account the nodes lying between subgraphs, as well as the relevance of these nodes to the overall subgraph interconnectivity. Experiments were carried out using four types of network models and five instances of real-world networks, in order to illustrate how subgraph characterization can help complementing complex network-based studies.

Feature selection environment for genomic applications

Background: Feature selection is a pattern recognition approach to choose important variables according to some criteria in order to distinguish or explain certain phenomena (i.e., for dimensionality reduction). There are many genomic and proteomic applications that rely on feature selection to answer questions such as selecting signature genes which are informative about some biological state, e. g., normal tissues and several types of cancer; or inferring a prediction network among elements such as genes, proteins and external stimuli. In these applications, a recurrent problem is the lack of samples to perform an adequate estimate of the joint probabilities between element states. A myriad of feature selection algorithms and criterion functions have been proposed, although it is difficult to point the best solution for each application. Results: The intent of this work is to provide an open-source multiplataform graphical environment for bioinformatics problems, which supports many feature selection algorithms, criterion functions and graphic visualization tools such as scatterplots, parallel coordinates and graphs. A feature selection approach for growing genetic networks from seed genes ( targets or predictors) is also implemented in the system. Conclusion: The proposed feature selection environment allows data analysis using several algorithms, criterion functions and graphic visualization tools. Our experiments have shown the software effectiveness in two distinct types of biological problems. Besides, the environment can be used in different pattern recognition applications, although the main concern regards bioinformatics tasks.

Relevance of leaf surface contamination for assessing chemical composition of bromeliads in a restinga forest

Resuspended soil and other airborne particles adhered to the leaf surface affect the chemical composition of the plant. A well-defined cleaning procedure is necessary to avoid this problem, providing a correct assessment of the inherent chemical composition of bromeliads. To evaluate the influence of a washing procedure, INAA was applied for determining chemical elements in the leaves of bromeliads from Vriesea carinata species, both non-washed and washed with Alconox, EDTA and bi-distilled water. Br, Ce, Hg, La, Sc, Se, Sm and Th showed higher mass fractions in non-washed leaves. The washing procedure removed the exogenous material without leaching chemical elements from inside the tissues.

Kernel machines for epilepsy diagnosis via EEG signal classification: A comparative study

Objective: We carry out a systematic assessment on a suite of kernel-based learning machines while coping with the task of epilepsy diagnosis through automatic electroencephalogram (EEG) signal classification. Methods and materials: The kernel machines investigated include the standard support vector machine (SVM), the least squares SVM, the Lagrangian SVM, the smooth SVM, the proximal SVM, and the relevance vector machine. An extensive series of experiments was conducted on publicly available data, whose clinical EEG recordings were obtained from five normal subjects and five epileptic patients. The performance levels delivered by the different kernel machines are contrasted in terms of the criteria of predictive accuracy, sensitivity to the kernel function/parameter value, and sensitivity to the type of features extracted from the signal. For this purpose, 26 values for the kernel parameter (radius) of two well-known kernel functions (namely. Gaussian and exponential radial basis functions) were considered as well as 21 types of features extracted from the EEG signal, including statistical values derived from the discrete wavelet transform, Lyapunov exponents, and combinations thereof. Results: We first quantitatively assess the impact of the choice of the wavelet basis on the quality of the features extracted. Four wavelet basis functions were considered in this study. Then, we provide the average accuracy (i.e., cross-validation error) values delivered by 252 kernel machine configurations; in particular, 40%/35% of the best-calibrated models of the standard and least squares SVMs reached 100% accuracy rate for the two kernel functions considered. Moreover, we show the sensitivity profiles exhibited by a large sample of the configurations whereby one can visually inspect their levels of sensitiveness to the type of feature and to the kernel function/parameter value. Conclusions: Overall, the results evidence that all kernel machines are competitive in terms of accuracy, with the standard and least squares SVMs prevailing more consistently. Moreover, the choice of the kernel function and parameter value as well as the choice of the feature extractor are critical decisions to be taken, albeit the choice of the wavelet family seems not to be so relevant. Also, the statistical values calculated over the Lyapunov exponents were good sources of signal representation, but not as informative as their wavelet counterparts. Finally, a typical sensitivity profile has emerged among all types of machines, involving some regions of stability separated by zones of sharp variation, with some kernel parameter values clearly associated with better accuracy rates (zones of optimality). (C) 2011 Elsevier B.V. All rights reserved.

Relevance of extractives and wood transformation products on the biodegradation of Pinus taeda by Ceriporiopsis subvermispora

Ceriporiopsis subvermispora is a promising white-rot fungus for biopulping. However, the underlying biochemistry involved in lignin removal and insignificant cellulose degradation by this species is not completely understood. This paper addresses this topic focusing on the involvement of ethanol-soluble extractives and wood transformation products in the biodegradation process. Cultures containing ethanol-extracted or in natura wood chips presented similar levels of extracellular enzymes and degradation of wood components. Fe3+-reducing compounds present in undecayed Pinus taeda were rapidly diminished by fungal degradation. Lignin-degradation products released during biodegradation restored part of the Fe3+-reducing activity. However, Fe3+ reduction was ineffective in presence of 0.5 mM oxalate at pH 4.5. Fungal consumption of Fe3+-reducing compounds and secretion of oxalic acid minimized the significance of Fenton`s reaction in the initial stages of wood biotreatment. This would explain limited polysaccharide degradation by the fungus that also lacks a complete set of hydrolytic enzymes. Scientific relevance of the paper: Ceriporiopsis subvermispora is a white-rot fungus suitable for biopulping processes because it degrades lignin selectively and causes significant structural changes on the wood components during the earlier decay stages. However, the intricate mechanism to explain lignin transformation and insignificant cellulose degradation by this species remains poorly understood. Some recent evidences pointed out for lipid peroxidation reactions as all initiating process explaining lignin degradation. On the other hand, alkylitaconic acids produced by the fungus via transformations of fatty acids occurring in wood showed to prevent polysaccharide degradation in Fenton reactions. In this context, one may conclude that the involvement of native wood substances or their transformation products in the overall wood biodegradation process induced by C subvermispora is still a matter of discussion. While free and esterified fatty acids present in wood extractives may be involved in the biosynthesis of alkylitaconic acids and in lipid peroxidation reactions, some extractives and lignin degradation products can reduce Fe3+, providing Fe2+ species needed to form OH radical via Fenton`s reaction. The present study focuses on this topic by evaluating the relevance of ethanol-soluble extractives and wood transformation products on the biodegradation of P. taeda by C subvermispora. For this, solid-state cultures containing ethanol-extracted and in natura wood chips were evaluated in details for up to 4 weeks. (C) 2007 Elsevier Ltd. All rights reserved.

Lignocellulosic polysaccharides and lignin degradation by wood decay fungi: the relevance of nonenzymatic Fenton-based reactions

The brown rot fungus Wolfiporia cocos and the selective white rot fungus Perenniporia medulla-panis produce peptides and phenolate-derivative compounds as low molecular weight Fe(3+)-reductants. Phenolates were the major compounds with Fe(3+)-reducing activity in both fungi and displayed Fe(3+)-reducing activity at pH 2.0 and 4.5 in the absence and presence of oxalic acid. The chemical structures of these compounds were identified. Together with Fe(3+) and H(2)O(2) (mediated Fenton reaction) they produced oxygen radicals that oxidized lignocellulosic polysaccharides and lignin extensively in vitro under conditions similar to those found in vivo. These results indicate that, in addition to the extensively studied Gloeophyllum trabeum-a model brown rot fungus-other brown rot fungi as well as selective white rot fungi, possess the means to promote Fenton chemistry to degrade cellulose and hemicellulose, and to modify lignin. Moreover, new information is provided, particularly regarding how lignin is attacked, and either repolymerized or solubilized depending on the type of fungal attack, and suggests a new pathway for selective white rot degradation of wood. The importance of Fenton reactions mediated by phenolates operating separately or synergistically with carbohydrate-degrading enzymes in brown rot fungi, and lignin-modifying enzymes in white rot fungi is discussed. This research improves our understanding of natural processes in carbon cycling in the environment, which may enable the exploration of novel methods for bioconversion of lignocellulose in the production of biofuels or polymers, in addition to the development of new and better ways to protect wood from degradation by microorganisms.

Distribution of transferrin saturation in an Australian population: Relevance to the early diagnosis of hemochromatosis

Background & Aims: An elevated transferrin saturation is the earliest phenotypic abnormality in hereditary hemochromatosis. Determination of transferrin saturation remains the most useful noninvasive screening test for affected individuals, but there is debate as to the appropriate screening level. The aims of this study were to estimate the mean transferrin saturation in hemochromatosis heterozygotes and normal individuals and to evaluate potential transferrin saturation screening levels. Methods: Statistical mixture modeling was applied to data from a survey of asymptomatic Australians to estimate the mean transferrin saturation in hemochromatosis heterozygotes and normal individuals. To evaluate potential transferrin saturation screening levels, modeling results were compared with data from identified hemochromatosis heterozygotes and homozygotes. Results: After removal of hemochromatosis homozygotes, two populations of transferrin saturation were identified in asymptomatic Australians (P < 0.01). In men, 88.2% of the truncated sample had a lower mean transferrin saturation of 24.1%, whereas 11.8% had an increased mean transferrin saturation of 37.3%. Similar results were found in women, A transferrin saturation threshold of 45% identified 98% of homozygotes without misidentifying any normal individuals. Conclusions: The results confirm that hemochromatosis heterozygotes form a distinct transferrin saturation subpopulation and support the use of transferrin saturation as an inexpensive screening test for hemochromatosis. In practice, a fasting transferrin saturation of greater than or equal to 45% identifies virtually all affected homozygous subjects without necessitating further investigation of unaffected normal individuals.

Initial results of ultrasound screening for aneurysm of the abdominal aorta in Western Australia: relevance for endoluminal treatment of aneurysm disease

Background. Increased life expectancy in men during the last thirty years is largely due to the decrease in mortality from cardiovascular disease in the age group 29-69 yr. This change has resulted in a change in the disease profile of the population with conditions such as aneurysm of the abdominal aorta (AAA) becoming more prevalent. The advent of endoluminal treatment for AAA has encouraged prophylactic intervention and fuelled the argument to screen for the disease. The feasibility of inserting an endoluminal graft is dependent on the morphology and growth characteristics of the aneurysm. This study used data from a randomized controlled trial of ultrasound screening for AAA in men aged 65-83 yr in Western Australia for the purpose of determining the norms of the living anatomy in the pressurized infrarenal aorta. Aims. To examine (1) the diameters of the infra-renal aorta in aneurysmal and non-aneurysmal cases, (2) the implications for treatment modalities, with particular reference to endoluminal grafting, which is most dependent on normal and aneurysmal morphology, and (3) any evidence to support the notion that northern Europeans are predisposed to aneurysmal disease. Methods. Using ultrasound, a randomized control trial was established in Western Australia to assess the value of a screening program in males aged 65-83 yr, The infra-renal aorta was defined as aneurysmal if the maximum diameter was 30 mm or more. Aortic diameter was modelled both as a continuous tin mm) and as a binary outcome variable, for those men who had an infra-renal diameter of 30 mm or more. ANOVA and linear regression were used for modelling aortic diameter as a continuum, while chi-square analysis and logistic regression were used in comparing men with and without the diagnosis of AAA. Findings. By December 1998, of 19.583 men had been invited to undergo ultrasound screening for AAA, 12.203 accepted the invitation (corrected response fraction 70.8%). The prevalence of AAA increased with age from 4.8% at 65 yr to 10.8% at 80 yr (chi (2) = 77.9, df = 3, P<0.001). The median (IQR) diameter for the non-aneurysmal group was 21.4 mm (3.3 mm) and there was an increase (<chi>(2) = 76.0, df = 1, P<0.001) in the diameter of the infra-renal aorta with age. Since 27 mm is the 95th centile for the non-aneurysmal infra-renal aorta, a diameter of 30 mm or more is justified as defining an aneurysm. The risk of AAA was higher in men of Australian (OR = 1.0) and northern European origin (OR = 1.0, 95%CL: 0.9. 1.2) compared with those of Mediterranean origin (OR = 0.5, 99%CL: 0.4, 0.7). Conclusion. Although screening has not yet been shown to reduce mortality from AAA. these population-based data assist the understanding of aneurysmal disease and the further development and use of endoluminal grafts for this condition. (C) 2001 Published by Elsevier Science Ltd on behalf of The International Society for Cardiovascular Surgery.

The Controllership Relevance or Risk Management in non-Financial Companies: a study of risk managers` and controllers` perceptions

The purpose of this study is to analyze the Controllership relevance as support risk management in non-financial companies. Risk management is a widely discussed and disseminated subject amongst financial institutions. It is obvious that economic uncertainties and, consequently, prevention and. control must also exist in non-financial companies. To enable managers to take safe-decisions, it is essential for them to be able to count on instrumental support that provides timely and adequate information, to ensure lower levels of mistakes and risk exposure. However, discussion concerning risk management in non-financial companies is still in its early stages in Brazil. Considering this gap, this study aims at assessing how Controllership has been acting in? companies under the insight of risk and how it can contribute to risk management in non-financial companies. To achieve the proposed goal, a field research was. carried-out with non-financial companies that are located in the city Sao Paulo and listed in the Sao Paulo Stock Exchange (Bovespa). The research was carried out using questionnaires, which were sent do Risk Officers and Controllers of those companies with the purpose of evaluating their perception on the subject. The results,of the research allow us to conclude that Controllership offers support to risk management, through information that contributes to the mitigation of the risks in non-financial companies.