Papers by ROMULO DE SOUSA
A spherical self-organizing map (SSOM) based on an icosahedral geodesic dome (ICOSOM) improves th... more A spherical self-organizing map (SSOM) based on an icosahedral geodesic dome (ICOSOM) improves the ability to visualize interactions among clusters from the input space. The SSOM reveals more information about the clusters’ properties than the original two-dimensional SOM (2D SOM) data maps, where clusters can position themselves at the edges. However, to completely visualize the spherical map, an ICOSOM requires a cumbersome data map projection using a virtual environment or cartographic projection that complicates the analyses of labels in the data map. The SSOM based on a tetrahedral geodesic dome (4HSOM) is flexible for sizing a lattice and enables the use of a more straightforward projection to obtain a data map with a complete view of the entire surface of a spherical lattice and a better analysis of the labels, such as 2D SOM projection. Nonetheless, the 4HSOM irregular lattice can interfere with the learning process and impair the visualization of the input space topographic...
Gestão das Águas na Amazônia: Atores Sociais, Marcos Regulatórios e Escalas
anppas.org.br
Page 1. IV Encontro Nacional da Anppas 4, 5 e 6 de junho de 2008 Brasília - DF Brasil _____ Ges... more Page 1. IV Encontro Nacional da Anppas 4, 5 e 6 de junho de 2008 Brasília - DF Brasil _____ Gestão das Águas na Amazônia: Atores Sociais, Marcos Regulatórios e Escalas ...
Anais do 10. Congresso Brasileiro de Inteligência Computacional, 2016
Resumo -Este trabalho propõe a utilização do sólido platônico tetraedro para projeção geodésica n... more Resumo -Este trabalho propõe a utilização do sólido platônico tetraedro para projeção geodésica na pavimentação de grade neural esférica de redes neurais artificiais (RNA) de Kohonen, ou mapas auto-organizáveis (self-organizing map -SOM). Grades neurais esféricas são utilizadas para eliminar o efeito de borda inerente às grades neurais definidas no espaço euclidiano. A projeção tetraédrica estabelece intervalos menores no crescimento de uma grade neural, diferentemente do que ocorre com a projeção do icosaedro, que é comumente utilizado por permitir uma topologia de rede da grade neural mais regular, por requerer uma menor variação no tamanho das cordas geodésicas projetadas na superfície da esfera. Contudo, o icosaedro requer um crescimento exponencial no dimensionamento das grades neurais, o que limita a processo de escolha da quantidade de neurônios que possibilite uma melhor qualidade na projeção e quantização vetorial do algoritmo SOM. Para apresentação dos resultados são utilizadas duas bases de dados, uma com pontos distribuídos de forma uniforme em um espaço euclidiano tridimensional, e outra com dados socioeconômicos de um projeto de pesquisa de ação antrópica sobre o meio ambiente.
A Política Das Águas Na Amazônia: As Especificidades Da Relação Entre O Marco Legal e Os Usuários Da Bacia Do Rio Purus
Optimization of Geodesic Self-Organizing Map
The 2012 International Joint Conference on Neural Networks (IJCNN), 2012
ABSTRACT The Geodesic Self-Organizing Map (GeoSOM) is a variation of traditional SOM, which uses ... more ABSTRACT The Geodesic Self-Organizing Map (GeoSOM) is a variation of traditional SOM, which uses an icosahedron-based tessellation as spherical lattice to eliminate the border effect to minimize the distortion in the reduction of high-dimensional spaces. Border effect is a problem intrinsic of low-dimensional neural grid, where neurons in the border have a less possibility to have its synaptic weights updated. The almost perfect regularity of a tessellated icosahedron projection onto a sphere solves this problem, reducing in two thirds of the distortion of 2D SOM. However, two problems appear resulting from complex shape of this Platonic polyhedron. First, the growth curve of lattice sizing follows a strong upward tendency that means a loss of control over the lattice sizing. Second, an overall visualization of topographic map is only possible with a geodesic projection of prototype vector positions from the surface of the sphere to a 2D plane that results in a elliptical map, flat on top and bottom, that avoids an orthogonal alignment of the data in the left and right sides, causing some distortion in the presentation of results and avoid an intuitive visualization of the map. This work proposes a geodesic self-organizing map, called 4HSOM, which uses a tessellated tetrahedron as lattice to eliminate the border effect, maximizing the control over the lattice sizing, with an easier overall visualization of topographic map without any geodetic projection, resulting from the minimalist geometric structure of the tetrahedron, although the tessellated tetrahedron has the greatest irregularity among the Platonic polyhedra. The work presents a comparative analysis between the results achieved by 4HSOM and GeoSOM.
Anais do 10. Congresso Brasileiro de Inteligência Computacional, 2016
This paper presents the results achieved by an application of Computational Intelligence techniqu... more This paper presents the results achieved by an application of Computational Intelligence technique in an Environmental and Social research activity in an innovative initiative, using a computational model formed by two Artificial Neural Networks (ANN) of distinct intentions to construct an environmental indicator and a characteristics map, to assist researchers and decision makers in the analysis of human intervention, through the measurement of the capacity to conserve the environment of a Amazonian community. The use of this computational model distinguishes itself in relation to the classic mathematical models, because it extracts the cognitive perceptions reached by researchers during the process of field research, avoiding distortions of the reality.
Uploads
Papers by ROMULO DE SOUSA