Estudo Comparativo de Métricas de Pontuação para Aprendizagem Estrutural de Redes Bayesianas

Bayesian networks are powerful tools as they represent probability distributions as graphs. They work with uncertainties of real systems. Since last decade there is a special interest in learning network structures from data. However learning the best network structure is a NP-Hard problem, so many heuristics algorithms to generate network structures from data were created. Many of these algorithms use score metrics to generate the network model. This thesis compare three of most used score metrics. The K-2 algorithm and two pattern benchmarks, ASIA and ALARM, were used to carry out the comparison. Results show that score metrics with hyperparameters that strength the tendency to select simpler network structures are better than score metrics with weaker tendency to select simpler network structures for both metrics (Heckerman-Geiger and modified MDL). Heckerman-Geiger Bayesian score metric works better than MDL with large datasets and MDL works better than Heckerman-Geiger with small datasets. The modified MDL gives similar results to Heckerman-Geiger for large datasets and close results to MDL for small datasets with stronger tendency to select simpler network structures

Preparo e caracterização de membranas de quitosana modificadas com poli (ácido acrílico)

The aim of this study was to generate an asymmetric biocompactible and biodegradable chitosan membrane modified by the contact with a poly(acrylic acid) solution at one of its sides at room temperature and 60◦C. The pure chitosan membrane, as well as the ones treated with poly(acrylic acid) were characterized by infrared spectroscopy (FTIRATR) at angles of 39◦, 45◦ and 60◦ , swelling capacity in water, thermal analysis (TG/DTG), scanning electronic microscopy (SEM) and permeation experiments using metronidazole at 0,1% and 0,2% as a model drug. The results confirmed the presence of ionic interaction between chitosan and poly(acrylic acid) by means of a polyelectrolyte complex (PEC) formation. They also showed that such interactions were more effective at 60◦C since this temperature is above the chitosan glass transition temperature wich makes the diffusion of poly(acrylic acid) easier, and that the two treated membranes were asymmetrics, more thermically stable and less permeable in relation to metronidazole than the pure chitosan membrane

Analisando o desempenho do ClassAge: um sistema multiagentes para classificação de padrões

The use of multi-agent systems for classification tasks has been proposed in order to overcome some drawbacks of multi-classifier systems and, as a consequence, to improve performance of such systems. As a result, the NeurAge system was proposed. This system is composed by several neural agents which communicate and negotiate a common result for the testing patterns. In the NeurAge system, a negotiation method is very important to the overall performance of the system since the agents need to reach and agreement about a problem when there is a conflict among the agents. This thesis presents an extensive analysis of the NeurAge System where it is used all kind of classifiers. This systems is now named ClassAge System. It is aimed to analyze the reaction of this system to some modifications in its topology and configuration

Espectros fractais em sistemas nanoestruturados e cristais fotônicos

The study of the elementary excitations such as photons, phonons, plasmons, polaritons, polarons, excitons and magnons, in crystalline solids and nanostructures systems are nowdays important active field for research works in solid state physics as well as in statistical physics. With this aim in mind, this work has two distinct parts. In the first one, we investigate the propagation of excitons polaritons in nanostructured periodic and quasiperiodic multilayers, from the description of the behavior for bulk and surface modes in their individual constituents. Through analytical, as well as computational numerical calculation, we obtain the spectra for both surface and bulk exciton-polaritons modes in the superstructures. Besides, we investigate also how the quasiperiodicity modifies the band structure related to the periodic case, stressing their amazing self-similar behavior leaving to their fractal/multifractal aspects. Afterwards, we present our results related to the so-called photonic crystals, the eletromagnetic analogue of the electronic crystalline structure. We consider periodic and quasiperiodic structures, in which one of their component presents a negative refractive index. This unusual optic characteristic is obtained when the electric permissivity and the magnetic permeability µ are both negatives for the same range of angular frequency ω of the incident wave. The given curves show how the transmission of the photon waves is modified, with a striking self-similar profile. Moreover, we analyze the modification of the usual Planck´s thermal spectrum when we use a quasiperiodic fotonic superlattice as a filter.

Síntese e caracterização de nanopartículas magnéticas de Co e Ni com aplicação em magneto hipertermia

Nanoparticles are importante for the study of new phenomena and for the development of new applications. Metallic magnetic nanoparticles like Cobalt and Nickel are important for their applications in nanoscience and nanotechnology. In this work, we report on the synthesis and characterization of Ni and Co nanoparticles. The nanoparticles were prepared by the modi- ed sol-gel method and were formed in the pore-network of the biopolymer quitosan. The reduction occurred in absence of H2 ux. The metallic particles and their monoxides have a face-centered- cubic structure. The metallic particles sizes ranged from 59 to 77 nm and from 19 to 50 nm for Ni and Co, respectively. Their monoxides chemically passivated the metallic cores, and after several weeks we have not observed further increase in oxidation. The synthesis method was tuned to obtain mainly the ferromagnetic phase. The system behaves like a core/shell structure with a ferromagnetic core and an antiferromagnetic shell. Exchange bias e ect was observed at temperatures below the Néel temperature. Both systems were submitted to an alternated magnetic eld and the heat released by the particles increased the temperature to 140°C in an interval of 5 min. Similar studies in samples dispersed in water increased the temperatures to 40-59°C, these results suggest that these materials are candidates for magnetic hyperthermia.