A survey of evolutionary algorithms for decision-tree induction

This paper presents a survey of evolutionary algorithms that are designed for decision-tree induction. In this context, most of the paper focuses on approaches that evolve decision trees as an alternate heuristics to the traditional top-down divide-and-conquer approach. Additionally, we present some alternative methods that make use of evolutionary algorithms to improve particular components of decision-tree classifiers. The paper's original contributions are the following. First, it provides an up-to-date overview that is fully focused on evolutionary algorithms and decision trees and does not concentrate on any specific evolutionary approach. Second, it provides a taxonomy, which addresses works that evolve decision trees and works that design decision-tree components by the use of evolutionary algorithms. Finally, a number of references are provided that describe applications of evolutionary algorithms for decision-tree induction in different domains. At the end of this paper, we address some important issues and open questions that can be the subject of future research.

Low Apgar scores at 5 minutes in a low risk population: maternal and obstetrical factors and postnatal outcome

Objective: To evaluate the association between Apgar scores of less than seven at five minutes (AS(5min) < 7) and antenatal factors and postnatal outcomes. Methods: A retrospective cohort and case-control study of 27,252 consecutive term newborns in a low risk obstetrical population between January 2003 and December 2010. Maternal and infant databases were reviewed from all cases with AS(5min) < 7 (n = 121; 0.4%) and 363 cases with AS(5min) >= 7 at 5 minutes who were randomly selected by a computer program. The main outcomes were neonatal death, newborn respiratory distress, need for orotracheal intubation and neonatal intensive care unit (NICU), and hypoxic-ischemic-encephalopathy. Results: After multiple regression analysis, repeated late decelerations on cardiotocography (OR: 2.4; 95% CI: 1.4-4.1) and prolonged second stage of labor (OR: 3.3; 95% CI: 1.3-8.3) were associated with AS(5min) < 7, as well as neonatal respiratory distress (OR: 3.0; 95% CI: 1.3-6.9), orotracheal intubation (OR: 2.5; 95% CI: 1.2-4.8), need for NICU (OR: 9.5; 95% CI: 6.7-16.8), and hypoxic-ischemic-encephalopathy (OR: 14.1; 95% CI: 3.6-54.7). No other antenatal factors were associated with AS(5min) < 7 (p > 0.05). Conclusion: Repeated late decelerations and prolonged second stage of labor in the low-risk population are predictors of AS(5min) < 7, a situation associated with increased risk of neonatal respiratory distress, need for mechanical ventilatory support and NICU, and hypoxic-ischemic-encephalopathy.

Cognitive function in patients with chronic pain treated with opioids: characteristics and associated factors

Background The paucity of studies regarding cognitive function in patients with chronic pain, and growing evidence regarding the cognitive effects of pain and opioids on cognitive function prompted us to assess cognition via neuropsychological measurement in patients with chronic non-cancer pain treated with opioids. Methods In this cross-sectional study, 49 patients were assessed by Continuous Reaction Time, Finger Tapping, Digit Span, Trail Making Test-B and Mini-mental State Examination tests. Linear regressions were applied. Results Patients scored poorly in the Trail Making Test-B (mean?=?107.6?s, SD?=?61.0, cut-off?=?91?s); and adequately on all other tests. Several associations among independent variables and cognitive tests were observed. In the multiple regression analyses, the variables associated with statistically significant poor cognitive performance were female sex, higher age, lower annual income, lower schooling, anxiety, depression, tiredness, lower opioid dose, and more than 5?h of sleep the night before assessment (P?<?0.05). Conclusions Patients with chronic pain may have cognitive dysfunction related to some reversible factors, which can be optimized by therapeutic interventions.

Estimativa de densidade amostral para elaboração de mapas de produtividade

Yield mapping represents the spatial variability concerning the features of a productive area and allows intervening on the next year production, for example, on a site-specific input application. The trial aimed at verifying the influence of a sampling density and the type of interpolator on yield mapping precision to be produced by a manual sampling of grains. This solution is usually adopted when a combine with yield monitor can not be used. An yield map was developed using data obtained from a combine equipped with yield monitor during corn harvesting. From this map, 84 sample grids were established and through three interpolators: inverse of square distance, inverse of distance and ordinary kriging, 252 yield maps were created. Then they were compared with the original one using the coefficient of relative deviation (CRD) and the kappa index. The loss regarding yield mapping information increased as the sampling density decreased. Besides, it was also dependent on the interpolation method used. A multiple regression model was adjusted to the variable CRD, according to the following variables: spatial variability index and sampling density. This model aimed at aiding the farmer to define the sampling density, thus, allowing to obtain the manual yield mapping, during eventual problems in the yield monitor.

Mapeamento de solos e da produtividade de plantações de Eucalyptus grandis, com uso de sistema de informação geográfica

Detailed environmental land characterization is essential for technical and financial planning, for both the scientific point of view and technological application. This work aimed at the physiographic and pedological characterization and eucalyptus productivity mapping at Itatinga Forest Sciences Experimental Station (southeastern Brazil), using geographic information systems in order to identify possible cause-effect relationships between forest productivity and soil attributes. The digital cartographic dataset was structured as follows: as primary source of data, aerial photograph and field survey were used and, as a secondary source, topographical, geological and land use occupation maps were used. For mapping wood productivity at age six (MAI6, Mean Annual Increment), inventory data of permanent plots (same species, provenance and age) were used, which were obtained from Eucalyptus grandis plantations. Using simple linear correlation and backward stepwise multiple regression analysis, the dependent variable (MAI) was related with physical and chemical characteristics of the soils. Two standards of contour curves were identified, one with close curves, narrow and surrounding the drainage network, in the steeper and lower altitude areas; the other, with spaced contour lines, in the areas of higher altitude and with plane relief. Six types of soils were characterized as being highly related to the physiographic patterns of the area: loamy sandy to sandy clayey Typic Hapludox (LVAd, 47.5%), clayey Rhodic Hapludox (LVd1, 33.4%), sandy clay Rhodic Hapludox (LVd2, 6%), clayey Rhodic Hapludox (LVdf, 9.1%), Entisols (G, 3.4%) and Fluvents soil (RY, 0.6%). There were large variations in wood productivity in the Eucalyptus grandis plantations, characterized in six classes, ranging from 26 to 52 m(3) ha(-1) yr(-1). These productivity changes were strictly related to soil mapping units. Through multiple regression analysis, we found that clay and organic matter contents were the attributes which most strongly explained the productivity differences.

Hantavirus pulmonary syndrome: prognostic factors for death in reported cases in Brazil

Hantavirus pulmonary syndrome (HPS) was described for the first time in Brazil in 1993 and has occurred endemically throughout the country. This study analysed clinical and laboratory aspects as well as death-related factors for HPS cases in Brazil from 1993 to 2006. The investigation comprised a descriptive and exploratory study of the history of cases as well as an analytical retrospective cohort survey to identify prognostic factors for death due to HPS. A total of 855 Brazilian HPS cases were assessed. The majority of cases occurred during spring (33.5%) and winter (27.6%), mainly among young male adults working in rural areas. The global case fatality rate was 39.3%. The mean interval between the onset of symptoms and hospitalisation was 4 days and that between hospitalisation and death was 1 day. In the multiple regression analysis, adult respiratory distress syndrome and mechanical respiratory support were associated with risk of death; when these two variables were excluded from the model, dyspnoea and haemoconcentration were associated with a higher risk of death. (C) 2012 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.

Estimating subsoil resistance to nitrate leaching from easily measurable pedological properties

Leaching of nitrate (NO3-) can increase the groundwater concentration of this anion and reduce the agronomical effectiveness of nitrogen fertilizers. The main soil property inversely related to NO3- leaching is the anion exchange capacity (AEC), whose determination is however too time-consuming for being carried out in soil testing laboratories. For this reason, this study evaluated if more easily measurable soil properties could be used to estimate the resistance of subsoils to NO3- leaching. Samples from the subsurface layer (20-40 cm) of 24 representative soils of São Paulo State were characterized for particle-size distribution and for chemical and electrochemical properties. The subsoil content of adsorbed NO3- was calculated from the difference between the NO3- contents extracted with 1 mol L-1 KCl and with water; furthermore, NO3- leaching was studied in miscible displacement experiments. The results of both adsorption and leaching experiments were consistent with the well-known role exerted by AEC on the nitrate behavior in weathered soils. Multiple regression analysis indicated that in subsoils with (i) low values of remaining phosphorus (Prem), (ii) low soil pH values measured in water (pH H2O), and (iii) high pH values measured in 1 moL L-1 KCl (pH KCl), the amounts of surface positive charges tend to be greater. For this reason, NO3- leaching tends to be slower in these subsoils, even under saturated flow condition.