Classificação de padrões através de um comitê de máquinas aprimorado por aprendizagem por reforço

Reinforcement learning is a machine learning technique that, although finding a large number of applications, maybe is yet to reach its full potential. One of the inadequately tested possibilities is the use of reinforcement learning in combination with other methods for the solution of pattern classification problems. It is well documented in the literature the problems that support vector machine ensembles face in terms of generalization capacity. Algorithms such as Adaboost do not deal appropriately with the imbalances that arise in those situations. Several alternatives have been proposed, with varying degrees of success. This dissertation presents a new approach to building committees of support vector machines. The presented algorithm combines Adaboost algorithm with a layer of reinforcement learning to adjust committee parameters in order to avoid that imbalances on the committee components affect the generalization performance of the final hypothesis. Comparisons were made with ensembles using and not using the reinforcement learning layer, testing benchmark data sets widely known in area of pattern classification

Algoritmos genéticos aplicados a um comitê de LS-SVM em problemas de classificação

The pattern classification is one of the machine learning subareas that has the most outstanding. Among the various approaches to solve pattern classification problems, the Support Vector Machines (SVM) receive great emphasis, due to its ease of use and good generalization performance. The Least Squares formulation of SVM (LS-SVM) finds the solution by solving a set of linear equations instead of quadratic programming implemented in SVM. The LS-SVMs provide some free parameters that have to be correctly chosen to achieve satisfactory results in a given task. Despite the LS-SVMs having high performance, lots of tools have been developed to improve them, mainly the development of new classifying methods and the employment of ensembles, in other words, a combination of several classifiers. In this work, our proposal is to use an ensemble and a Genetic Algorithm (GA), search algorithm based on the evolution of species, to enhance the LSSVM classification. In the construction of this ensemble, we use a random selection of attributes of the original problem, which it splits the original problem into smaller ones where each classifier will act. So, we apply a genetic algorithm to find effective values of the LS-SVM parameters and also to find a weight vector, measuring the importance of each machine in the final classification. Finally, the final classification is obtained by a linear combination of the decision values of the LS-SVMs with the weight vector. We used several classification problems, taken as benchmarks to evaluate the performance of the algorithm and compared the results with other classifiers

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

Aninhamento em redes bipartidas

We present a nestedness index that measures the nestedness pattern of bipartite networks, a problem that arises in theoretical ecology. Our measure is derived using the sum of distances of the occupied elements in the adjacency matrix of the network. This index quantifies directly the deviation of a given matrix from the nested pattern. In the most simple case the distance of the matrix element ai,j is di,j = i+j, the Manhattan distance. A generic distance is obtained as di,j = (i¬ + j¬)1/¬. The nestedness índex is defined by = 1 − where is the temperature of the matrix. We construct the temperature index using two benchmarks: the distance of the complete nested matrix that corresponds to zero temperature and the distance of the average random matrix that is defined as temperature one. We discuss an important feature of the problem: matrix occupancy. We address this question using a metric index ¬ that adjusts for matrix occupancy

Proposta e implementação de uma arquitetura reconfigurável híbrida para aplicações baseadas em fluxo de dados

The increase of applications complexity has demanded hardware even more flexible and able to achieve higher performance. Traditional hardware solutions have not been successful in providing these applications constraints. General purpose processors have inherent flexibility, since they perform several tasks, however, they can not reach high performance when compared to application-specific devices. Moreover, since application-specific devices perform only few tasks, they achieve high performance, although they have less flexibility. Reconfigurable architectures emerged as an alternative to traditional approaches and have become an area of rising interest over the last decades. The purpose of this new paradigm is to modify the device s behavior according to the application. Thus, it is possible to balance flexibility and performance and also to attend the applications constraints. This work presents the design and implementation of a coarse grained hybrid reconfigurable architecture to stream-based applications. The architecture, named RoSA, consists of a reconfigurable logic attached to a processor. Its goal is to exploit the instruction level parallelism from intensive data-flow applications to accelerate the application s execution on the reconfigurable logic. The instruction level parallelism extraction is done at compile time, thus, this work also presents an optimization phase to the RoSA architecture to be included in the GCC compiler. To design the architecture, this work also presents a methodology based on hardware reuse of datapaths, named RoSE. RoSE aims to visualize the reconfigurable units through reusability levels, which provides area saving and datapath simplification. The architecture presented was implemented in hardware description language (VHDL). It was validated through simulations and prototyping. To characterize performance analysis some benchmarks were used and they demonstrated a speedup of 11x on the execution of some applications

Determinantes de competitividade de destinos turísticos aplicados aos roteiros regionais: uma avaliação do roteiro Seridó/RN

The tourism industry is gaining representation by move and stimulate the economy, especially by allowing the generation of employment and income, thus allowing growth opportunities for localities where tourism develops. Therefore, the present study entitled determinants of competitiveness of tourist destinations applied to regional routes: an evaluation of the Route of Seridó/RN, discusses the issue of competitiveness in tourism and tries to understand the scenery of this Route. The main objective of the study is to assess the conditions of competitiveness in the Route of Seridó/RN according to benchmarks and global determinants of competitiveness for tourist destinations. The study has also as specifics objectives: define dimensions of the reference model for use in evaluating the competitiveness of the Route of Seridó/RN; identify levels of governance and competitiveness in the municipalities that make up the sample set above the Route, and analyze to what extent the competitiveness of the Route correspond to the global reference of competitiveness of tourism destinations. Regarding the methodology, it is a search for an exploratory- descriptive and used a combination of quantitative and qualitative research method as expected and required in the implementation of the evaluation tool called Compet&enible Model. For data collection, it has been taken technical visits and also analysis of documents and materials. Data analysis was based on the records and documents and the use of simple descriptive statistics for the scores of the elements offered by Compet&enible Model. The results allowed us to know the real conditions of competitiveness of the Seridó/RN Route forward to the attributes of tourist destinations for global competitiveness: the dimension I, Governance, reached 17 points, classified as "in structuring" and dimension II, Competitiveness, reached 10 points, ranking "weak". These results highlight the need for greater involvement of the actors in the supply chain of tourism in Polo Seridó/RN for the actions, programs and projects are put into practice. It is expected that tourism is considered an important activity for the local and global development, serving as a reference for the future management of Seridó/RN Route, guiding new policy guidelines, planning and organization to better competitiveness

Terra e trabalho: permanência de uma política pública na Fazenda Mundo Novo, em Caicó-RN

This is study about the Project Vazantes, as a case of stay policy on access to land upstream of the weir humid New World, in Caicó-RN. It includes social and economic aspects of vazanteiros, that in their experiences with the land, recreate structures, forms and functions. Theoretically, the work is based on the ideas miltonianas (1986, 1988, 1996, 1997) on the production and reproduction of space, as seen this system of actions that is articulated to the public policies and the achievements solidarity, to the extent that production, in vazantes there is virtually no technology investment, technical assistance and rural credit. Regarding the search procedures, we conducted interviews and questionnaires with vazanteiros applied but did photographic records and research into documentary sources Agricultural Research Company's Rio Grande do Norte S / A (EMPARN), the Union of Rural Workers of Caicó (STRC ), In newspapers and magazines, as well as other sources of information related to the object of study.Such procedures, coupled with theoretical and methodological benchmarks, helped in the understanding of the relationship that takes place in the New World Farm, in order that space produced by actors (EMPARN and vazanteiros) comprises a structure of power that involves multiple interests in extent that land and labor are appropriate mechanisms as dialetizantes

Pedagogia do esporte: livro didático aplicado aos jogos esportivos coletivos

The study deals with the pedagogical possibilities of text books as a resource for Physical Education (PE) classes based on sports, specifically team sports. It started from a bibliographical review on the cited terms and after that we advanced our studies in order to propose a didactic text book on Team Sports for PE classes considering two benchmarks: one technical-tactic and the other social-educational. In order to do so, we used methods from linguistics, and the development of a Didactic Sequence allowed the selection and distribution of the contents to be presented in the book in an orderly and judicious way. The focus in Team Sports is justified by its social-cultural insertion and by the need to transform recent studies into new pedagogical procedures for practicing them. The didactic text book, as well as the adequate preparation of the professionals who will use it, is pointed out as facilitator in the process of teaching, experiencing and learning team sports.

Search for high-mass resonances decaying into tau-lepton pairs in pp collisions at root s=7 TeV

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Search for Resonant WW and WZ Production in p(p)over-bar Collisions at root s=1.96 TeV

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Antioxidant activity of basil and oregano extracts added to soybean oil for accelerated storage test

The objective of this research was to evaluate the antioxidant activity of extract of basil and oregano, and its synergistic effect when added to soybean oil and subjected to accelerated storage test. Extracts of basil, oregano and mixtures of both were applied to soybean oil at a concentration of 2.000mg/kg, then the extracts were heated in an oven at 60C for a period of 10 days. Samples were taken every 2 days and analyzed for concentrations of peroxides and conjugated dienes. Synthetic antioxidant tert-butylhydroquinone (TBHQ) at a concentration of 50mg/kg and soybean oil free antioxidant (control) subject to the same conditions were used as benchmarks. The results showed that the antioxidant effect decreases according to oil heating for all treatments. However, the synthetic antioxidant showed superior protection to the soybean oil during the formation of primary oxidation compounds, followed by the natural extracts, which showed no synergism. © 2012 Wiley Periodicals, Inc.

Nivelamento através do GPS: avaliação e proposição de estratégias

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Implementação e teste do método da fronteira imersa para a simulação do escoamento em torno de cilindros estacionários e rotativos

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Trabalho colaborativo entre universidade e escola: proporcionando a formação continuada através da história da biologia

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Análise de linguagens de modelagem de processos de um modelo de referência na cadeia de suprimentos

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)