Use of computational intelligence in the genetic divergence of colored cotton plants

The objective of this work was to analyze the genetic diversity using conventional methods and artificial neural networks among 12 colored fiber cotton genotypes, using technological characteristics of the fiber and productivity in terms of cottonseed and cotton fiber yield. The experiment was conducted in an experimental area located at Fazenda Capim Branco, belonging to the Federal University of Uberlândia, in the city of Uberlândia, Minas Gerais. Twelve genotypes of colored fiber cotton were evaluated, 10 from the Cotton Genetic Improvement Program (PROMALG): UFUJP - 01, UFUJP - 02, UFUJP - 05, UFUJP - 08, UFUJP - 09, UFUJP - 10, UFUJP - 11, UFUJP - 13, UFUJP - 16, UFUJP - 17 and two commercial cultivars: BRS Rubi (RC) and BRS Topázio (TC). The experimental design used was complete randomized block (CRB) with three replications. The following evaluations were carried out at full maturation: yield of cottonseed (kg ha-1) and the technological characteristics, which include, fiber length, micronaire, maturation, length uniformity, short fiber index, elongation and strength, using the HVI (High volume instrument) device. Genetic dissimilarity was measured using the generalized Mahalanobis distance and after obtaining the dissimilarity matrix, the genotypes were grouped using a hierarchical clustering method (UPGMA). A discriminant analysis and the Kohonen Self-Organizing Map (SOM) by Artificial Neural Networks (ANN's) were performed through computational intelligence. SOM was able to detect differences and organize the similarities between accesses in a more coherent way, forming a larger number of groups, when compared to the method that uses the Mahalanobis matrix. It was also more accurate than the discriminant analysis, since it made it possible to differentiate groups more coherently when comparing their phenotypic behavior. The methods that use computational intelligence proved to be more efficient in detecting similarity, with Kohonen's Self-Organizing Map being the most adequate to classify and group cotton genotypes.

Mapeamento dos aspectos hidrogeoquímicos de águas subterrâneas a partir de estatística multivariada e redes neurais artificiais / Mapping of groundwater hydrogeochemistry aspects through multivariate statistics and artificial neural networks

RESUMO O objetivo central deste trabalho foi realizar o mapeamento dos aspectos hidrogeoquímicos de águas subterrâneas usando a estatística multivariada e redes neurais artificiais como subsídio para identificação de padrões espaciais. Para tal, foi executado um estudo de caso em aquíferos no município de Lençóis, Bahia, na região da Chapada Diamantina, nordeste do Brasil. Foram realizadas campanhas de campo para coleta de coordenadas geodésicas e amostras de águas subterrâneas. Após análise laboratorial e determinação de dados analíticos, foi feita a interpretação dos processos ambientais com o uso da análise de agrupamentos e mapas auto-organizáveis, além de classificação das águas pela Resolução do Conselho Nacional do Meio Ambiente nº 396/2008. Para fins de mapeamento dos dados analisados, foram usadas técnicas de geoprocessamento no Sistema de Informação Geográfica. Os principais constituintes físicos e químicos analisados em dois períodos climáticos foram mapeados e divididos em sete agrupamentos. Foram identificadas quatro zonas no município, que apresentam diferentes contextos hidrogeoquímicos. As zonas dos setores leste/sudeste, sul (área urbana) e extremo sul apresentam as mais significativas alterações na hidrogeoquímica e qualidade das águas. O mapeamento, subsidiado pela estatística multivariada e redes neurais artificiais, se apresentou potencialmente útil em contribuir com as ações de gestão dos recursos hídricos subterrâneos, como delimitação de áreas prioritárias, monitoramento de zonas de riscos de contaminação, além de intervenções de engenharia que eventualmente busquem o saneamento ambiental das águas subterrâneas.

ABSTRACT The main objective this paper was to map the hydrogeochemistry aspects of groundwater using multivariate statistics and artificial neural networks as a subsidy to identify spatial patterns. For this, a case study was carried out in aquifers in the municipality of Lençóis (BA), in the region of Chapada Diamantina, Northeastern Brazil. Field campaigns were carried out to collect geodetic coordinates and groundwater samples. After laboratorial analysis and determination of analytical data, the environmental processes were interpreted by cluster analysis and self-organizing maps, as well as the waters classification through CONAMA Resolution no. 396/2008. For the purpose of mapping the analyzed data, geoprocessing techniques were used in GIS. The main physical and chemical constituents analyzed in two climatic periods were mapped and divided into seven clusters. Four zones that present different hydrogeochemical contexts were identified in the municipality. The zones of the east/southeastern, south (urban area) and south end sectors present the most significant changes in hydrogeochemistry and water quality. The mapping, supported by multivariate statistics and artificial neural networks, was potentially useful in contributing to the management actions of groundwater resources as delimitation of priority areas, monitoring of contamination risk zones and engineering interventions that eventually seek environmental groundwater sanitation.

Performance quantification of clustering algorithms for false positive removal in fMRI by ROC curves

Abstract Introduction Functional magnetic resonance imaging (fMRI) is a non-invasive technique that allows the detection of specific cerebral functions in humans based on hemodynamic changes. The contrast changes are about 5%, making visual inspection impossible. Thus, statistic strategies are applied to infer which brain region is engaged in a task. However, the traditional methods like general linear model and cross-correlation utilize voxel-wise calculation, introducing a lot of false-positive data. So, in this work we tested post-processing cluster algorithms to diminish the false-positives. Methods In this study, three clustering algorithms (the hierarchical cluster, k-means and self-organizing maps) were tested and compared for false-positive removal in the post-processing of cross-correlation analyses. Results Our results showed that the hierarchical cluster presented the best performance to remove the false positives in fMRI, being 2.3 times more accurate than k-means, and 1.9 times more accurate than self-organizing maps. Conclusion The hierarchical cluster presented the best performance in false-positive removal because it uses the inconsistency coefficient threshold, while k-means and self-organizing maps utilize a priori cluster number (centroids and neurons number); thus, the hierarchical cluster avoids clustering scattered voxels, as the inconsistency coefficient threshold allows only the voxels to be clustered that are at a minimum distance to some cluster.