Uma contribuição ao estudo das categorias internas e de sua proliferação em redes ARTMAP

ART networks present some advantages: online learning; convergence in a few epochs of training; incremental learning, etc. Even though, some problems exist, such as: categories proliferation, sensitivity to the presentation order of training patterns, the choice of a good vigilance parameter, etc. Among the problems, the most important is the category proliferation that is probably the most critical. This problem makes the network create too many categories, consuming resources to store unnecessarily a large number of categories, impacting negatively or even making the processing time unfeasible, without contributing to the quality of the representation problem, i. e., in many cases, the excessive amount of categories generated by ART networks makes the quality of generation inferior to the one it could reach. Another factor that leads to the category proliferation of ART networks is the difficulty of approximating regions that have non-rectangular geometry, causing a generalization inferior to the one obtained by other methods of classification. From the observation of these problems, three methodologies were proposed, being two of them focused on using a most flexible geometry than the one used by traditional ART networks, which minimize the problem of categories proliferation. The third methodology minimizes the problem of the presentation order of training patterns. To validate these new approaches, many tests were performed, where these results demonstrate that these new methodologies can improve the quality of generalization for ART networks

Identificação não linear usando uma rede fuzzy wavelet neural network modificada

In last decades, neural networks have been established as a major tool for the identification of nonlinear systems. Among the various types of networks used in identification, one that can be highlighted is the wavelet neural network (WNN). This network combines the characteristics of wavelet multiresolution theory with learning ability and generalization of neural networks usually, providing more accurate models than those ones obtained by traditional networks. An extension of WNN networks is to combine the neuro-fuzzy ANFIS (Adaptive Network Based Fuzzy Inference System) structure with wavelets, leading to generate the Fuzzy Wavelet Neural Network - FWNN structure. This network is very similar to ANFIS networks, with the difference that traditional polynomials present in consequent of this network are replaced by WNN networks. This paper proposes the identification of nonlinear dynamical systems from a network FWNN modified. In the proposed structure, functions only wavelets are used in the consequent. Thus, it is possible to obtain a simplification of the structure, reducing the number of adjustable parameters of the network. To evaluate the performance of network FWNN with this modification, an analysis of network performance is made, verifying advantages, disadvantages and cost effectiveness when compared to other existing FWNN structures in literature. The evaluations are carried out via the identification of two simulated systems traditionally found in the literature and a real nonlinear system, consisting of a nonlinear multi section tank. Finally, the network is used to infer values of temperature and humidity inside of a neonatal incubator. The execution of such analyzes is based on various criteria, like: mean squared error, number of training epochs, number of adjustable parameters, the variation of the mean square error, among others. The results found show the generalization ability of the modified structure, despite the simplification performed

Morfologia em alta resolução espacial de estrelas de tipo Mira e R Corenae Borealis

Mira and R Coronae Borealis (R CrB) variable stars are evolved objects surrounded by circumstellar envelopes (CSE) composed of the ejected stellar material. We present a detailed high-spatial resolution morfological study of the CSE of three stars: IRC+10216, the closest and more studied Carbon-Rich Mira; o Ceti, the prototype of the Mira class; and RY Sagitarii (RY Sgr), the brightest R CrB variable of the south hemisphere. JHKL near-infrared adaptive optics images of IRC+10216 with high dynamic range and Vband images with high angular resolution and high depth, collected with the VLT/NACO and VLT/FORS1 instruments, were analyzed. NACO images of o Ceti were also analyzed. Interferometric observations of RY Sgr collected with the VLTI/MIDI instrument allowed us to explore its CSE innermost regions (»20 40 mas). The CSE of IRC+10216 exhibit, in near-infrared, clumps with more complex relative displacements than proposed in previous studies. In V-band, the majority of the non-concentric shells, located in the outer CSE layers, seem to be composed of thinner elongated shells. In a global view, the morphological connection between the shells and the bipolar core of the nebulae, located in the outer layers, together with the clumps, located in the innermost regions, has a difficult interpretation. In the CSE of o Ceti, preliminar results would be indicating the presence of possible clumps. In the innermost regions (.110 UA) of the CSE of RY Sgr, two clouds were detected in different epochs, embedded in a variable gaussian envelope. Based on a rigorous verification, the first cloud was located at »100 R¤ (or »30 AU) from the centre, toward the east-north-east direction (modulo 180o) and the second one was almost at a perpendicular direction, having aproximately 2£ the distance of the first cloud. This study introduces new constraints to the mass-loss history of these kind of variables and to the morphology of their innermost CSE regions

Práticas agroecológicas para o cultivo do gergelim (Sesamum indicum L.)

The exceeded use of the natural resources required by the modern agriculture has been caused soil impoverishment, soil salinization and soil compaction. The unreasonable use of chemical fertilizers and pesticides causes chemical imbalances in the plant tissues, nutritional losses, taste chances and human health problems. The monocrops are more vulnerable to the pest and disease attacks. The aim of this work were to indicate the better relative planting time of the cowpea bean for the sesame/cowpea bean intercropping based on the evaluation of agronomic, economic, ecological and physiological parameters and to evaluate the viability of the cow urine and cassava wastewater use as alternative fertilizers on the sesame cultivation. In a field essay, when the sesame/cowpea bean was evaluate, the mainly treatments were the single planting of the two cultures and the multiple cropping, with the cowpea bean being planted on the same time, 7, 14 and 21 days after the sesame. The mainly treatments of the second essay were the combined doses of the cow urine (0 and 50 ml) and cassava wastewater (0, 500 and 1000 mL). In the intercrop evaluation significant differences were found between the treatments for the most of the parameters. It was verified that as the sowing of the cowpea bean was retarded in relation to the sesame sowing, the sesame performance increased when the cowpea bean performance decreased. The treatment 7 days balanced the competition relations that occurred between the two crops in the intercrop. The sesame presented little pronounced responses to the different doses of cassava wastewater. The cow urine affected the sesame growth and the growth rates increased with the increasing of the application of the product. The responses of the sesame growth to the cow urine application must be related both the nutritional richness of the product and the presence of growth stimulating substances. The obtained results indicated that in the sesame/cowpea bean intercropping, when the sesame is planted 7 days after the sesame, there is a higher possibility of the net gains to the farmer. Cow urine and cassava wastewater may be used as additional organic fertilizers

Consórcio de cártamo e feijão caupi: alternativa para a produção de biodiesel na agricultura familiar

The new development strategies should operate mainly in the areas of energy efficiency and sustainable agriculture. Thus, the substitution of fossil fuels with biofuels, such as biodiesel, is increasingly on the agenda. The cultivation of oilseed plants for biodiesel production must take place in integrated systems that enable best environmental benefits and are more economically significant. The objectives of this study were to assess the morphological, anatomic, and physiological characteristics of safflower (Carthamus tinctorius L., promising oilseed for biodiesel production) grown in monoculture and intercropping with cowpea bean (Vigna unguiculata L. Walp.); and identify socioeconomic family farmers and verify their acceptance about safflower as an energy crop. The methodology used for the analysis of safflower growth in monoculture and intercropped with beans, were morphoanatomical and histochemical analyzes, made with samples of plants grown in the field in two cropping systems throughout the range of the life cycle of these plants. There were no changes in growth and anatomy of plants, even in the consortium, which is satisfactory to indicate the intercropping system for those crops and can be a good alternative for the family farmer, who may have safflower as a source of income without giving up planting their livelihood. To check the acceptance of safflower by farmers, interviews were made to family farmers by Canudos agrovila in Ceará-Mirim/RN. It was noticed that many of them accept the introduction of safflower as oil crop, although unaware of the species, and that, being more resistant to drought, safflower help in the stability of families who depend on the weather conditions for success their current crops. In general, it is concluded that safflower has features that allows it to be grown in consortium for biodiesel production combined with the production of food, such as cowpea, and can be used enabling better development for family farmers.