Search

Dengue is considered one of the diseases with the most significant rates in Brazil. The increasing rates directly affect public health services, so that evaluating the environmental and social conditions in areas with high rates of the disease can assist in the development of diagnoses and health actions. In this sense, the objective of this study was to identify the most important socioeconomic variables for the prediction of dengue prevalence rates in municipalities of the state of Goiás. The evaluation was performed based on 38 socioeconomic variables obtained from the database of Instituto Brasileiro de Geografia e Estatística - IBGE, Fundação João Pinheiro - FJP and from the calculation of dengue prevalence rates based on data available in Sistema de Informação de Agravos de Notificação - SINAN for the periods 2001-2009 and 2010-2018. Modeling was performed from the evaluation of three machine learning algorithms: Random Forest, XGBoost and KNN. The results indicated that the most important variables showed an inverse relationship to the conditions of low income, illiteracy and deficiency in basic sanitation services. [ABSTRACT FROM AUTHOR]

This study aimed to test taper functions and artificial intelligence (AI) models in order to estimate merchantable volumes of Japanese cedar (Cryptomeria japonica) trees in a homogenous plantation in southern Brazil. A total of 30 individuals were rigorously scaled and their total volumes were calculated, including those of the following log assortments: veneer, sawn, pulp and energy. Three AI models, i.e. two variants ofk-nearest neighbours (KNN) instance-based classification (one and three nearest neighbours) and an artificial neural network (ANN) approach, were compared with three traditional taper models: fifth-order polynomial, fractional powers and the Garay model. The estimated volumes were compared with the actual volumes by means of the standard error (Syx), bias, precision and accuracy. Total volume estimates proved to be unbiased (maximum bias 5.42%), precise (maximum precision 9.28%) and accurate (maximum accuracy 10.79%) with all of the investigated models. The tested models tended to give lower bias, better precision and accuracy in the middle portion of the stems, but worse estimates at the base and tip (maximum bias −12.41%). In general, the KNN models improved merchantable volume estimation, particularly KNN1, which is a straightforward and simple method. We conclude that AI techniques have appeal for application in forest inventories and that KNN is a particularly interesting alternative for tree volume estimation. [ABSTRACT FROM PUBLISHER]

Abstract: The current legislation requires the mandatory addition of biodiesel to all Brazilian road diesel oil A (pure diesel) marketed in the country and bans the addition of vegetable oils for this type of diesel. However, cases of irregular addition of vegetable oils directly to the diesel oil may occur, mainly due to the lower cost of these raw materials compared to the final product, biodiesel. In Brazil, the situation is even more critical once the country is one of the largest producers of oleaginous products in the world, especially soybean, and also it has an extensive road network dependent on diesel. Therefore, alternatives to control the quality of diesel have become increasingly necessary. This study proposes an analytical methodology for quality control of diesel with intention to identify and determine adulterations of oils and even fats of vegetable origin. This methodology is based on detection, identification and quantification of triacylglycerols on diesel (main constituents of vegetable oils and fats) by high performance liquid chromatography in reversed phase with UV detection at 205nm associated with multivariate methods. Six different types of oils and fats were studied (soybean, frying oil, corn, cotton, palm oil and babassu) and two methods were developed for data analysis. The first one, based on principal component analysis (PCA), nearest neighbor classification (KNN) and univariate regression, was used for samples adulterated with a single type of oil or fat. In the second method, partial least square regression (PLS) was used for the cases where the adulterants were mixtures of up to three types of oils or fats. In the first method, the techniques of PCA and KNN were correctly classified as 17 out of 18 validation samples on the type of oil or fat present. The concentrations estimated for adulterants showed good agreement with the reference values, with mean errors of prediction (RMSEP) ranging between 0.10 and 0.22% (v/v). The PLS method was efficient in the quantification of mixtures of up to three types of oils and fats, with RMSEP being obtained between 0.08 and 0.27% (v/v), mean precision between 0.07 and 0.32% (v/v) and minimum detectable concentration between 0.23 and 0.81% (v/v) depending on the type of oil or fat in the mixture determined. [Copyright &y& Elsevier]

Abstract: This work describes using 1H NMR data and pattern recognition analysis to classify vinegars. Vinegar authenticity is linked to raw ingredient source and manufacturing conditions. Application of PCA and HCA methods resulted in the natural clustering of the samples according to the raw material used. Wine vinegars were characterized by a high concentration of ethyl acetate, glycerol, methanol and tartaric acid, while glycerol and ethyl acetate signals were not visible in alcohol/agrin vinegars. Apple vinegars showed to be richer in alanine. The KNN, SIMCA and PLS-DA methods were used to build predictive models for classification of vinegar type wine, apple and alcohol/agrin (27 samples – 22 as training set). The models were tested using an independent set (5 samples), no samples were wrongly classified. Validated models were used to predict the class of 21 commercial samples, which, as expected, were correctly classified. Eight commercial vinegars (honey, orange, pineapple and rice) were discriminated from these samples using PCA method. Honey vinegars did not present ethanol signals and pineapple vinegars presented the largest amount of tartaric acid. Rice and orange vinegars are richer in lactic acid and did not present the methanol signal. Alanine signals were not visible in orange vinegars. [Copyright &y& Elsevier]