Otimização em comitês de classificadores: uma abordagem baseada em filtro para seleção de subconjuntos de atributos

Traditional applications of feature selection in areas such as data mining, machine learning and pattern recognition aim to improve the accuracy and to reduce the computational cost of the model. It is done through the removal of redundant, irrelevant or noisy data, finding a representative subset of data that reduces its dimensionality without loss of performance. With the development of research in ensemble of classifiers and the verification that this type of model has better performance than the individual models, if the base classifiers are diverse, comes a new field of application to the research of feature selection. In this new field, it is desired to find diverse subsets of features for the construction of base classifiers for the ensemble systems. This work proposes an approach that maximizes the diversity of the ensembles by selecting subsets of features using a model independent of the learning algorithm and with low computational cost. This is done using bio-inspired metaheuristics with evaluation filter-based criteria

Uma nova forma de calcular os centros dos Clusters em algoritmos de agrupamento tipo fuzzy c-means

Clustering data is a very important task in data mining, image processing and pattern recognition problems. One of the most popular clustering algorithms is the Fuzzy C-Means (FCM). This thesis proposes to implement a new way of calculating the cluster centers in the procedure of FCM algorithm which are called ckMeans, and in some variants of FCM, in particular, here we apply it for those variants that use other distances. The goal of this change is to reduce the number of iterations and processing time of these algorithms without affecting the quality of the partition, or even to improve the number of correct classifications in some cases. Also, we developed an algorithm based on ckMeans to manipulate interval data considering interval membership degrees. This algorithm allows the representation of data without converting interval data into punctual ones, as it happens to other extensions of FCM that deal with interval data. In order to validate the proposed methodologies it was made a comparison between a clustering for ckMeans, K-Means and FCM algorithms (since the algorithm proposed in this paper to calculate the centers is similar to the K-Means) considering three different distances. We used several known databases. In this case, the results of Interval ckMeans were compared with the results of other clustering algorithms when applied to an interval database with minimum and maximum temperature of the month for a given year, referring to 37 cities distributed across continents

Contribuições para verificação automática de applets javacard

The widespread growth in the use of smart cards (by banks, transport services, and cell phones, etc) has brought an important fact that must be addressed: the need of tools that can be used to verify such cards, so to guarantee the correctness of their software. As the vast majority of cards that are being developed nowadays use the JavaCard technology as they software layer, the use of the Java Modeling Language (JML) to specify their programs appear as a natural solution. JML is a formal language tailored to Java. It has been inspired by methodologies from Larch and Eiffel, and has been widely adopted as the de facto language when dealing with specification of any Java related program. Various tools that make use of JML have already been developed, covering a wide range of functionalities, such as run time and static checking. But the tools existent so far for static checking are not fully automated, and, those that are, do not offer an adequate level of soundness and completeness. Our objective is to contribute to a series of techniques, that can be used to accomplish a fully automated and confident verification of JavaCard applets. In this work we present the first steps to this. With the use of a software platform comprised by Krakatoa, Why and haRVey, we developed a set of techniques to reduce the size of the theory necessary to verify the specifications. Such techniques have yielded very good results, with gains of almost 100% in all tested cases, and has proved as a valuable technique to be used, not only in this, but in most real world problems related to automatic verification

Novos métodos determinísticos para gerar centros iniciais dos grupos no algoritmo fuzzy C-Means e variantes

Data clustering is applied to various fields such as data mining, image processing and pattern recognition technique. Clustering algorithms splits a data set into clusters such that elements within the same cluster have a high degree of similarity, while elements belonging to different clusters have a high degree of dissimilarity. The Fuzzy C-Means Algorithm (FCM) is a fuzzy clustering algorithm most used and discussed in the literature. The performance of the FCM is strongly affected by the selection of the initial centers of the clusters. Therefore, the choice of a good set of initial cluster centers is very important for the performance of the algorithm. However, in FCM, the choice of initial centers is made randomly, making it difficult to find a good set. This paper proposes three new methods to obtain initial cluster centers, deterministically, the FCM algorithm, and can also be used in variants of the FCM. In this work these initialization methods were applied in variant ckMeans.With the proposed methods, we intend to obtain a set of initial centers which are close to the real cluster centers. With these new approaches startup if you want to reduce the number of iterations to converge these algorithms and processing time without affecting the quality of the cluster or even improve the quality in some cases. Accordingly, cluster validation indices were used to measure the quality of the clusters obtained by the modified FCM and ckMeans algorithms with the proposed initialization methods when applied to various data sets

Genética e biologia reprodutiva de vriesea minarum (Bromeliaceae): em busca de estratégias de conservação no quadrilátero ferrífero, minas gerais

Vriesea minarum is a rupiculous bromeliad species, with naturally fragmented populations, restricted to the Iron Quadrangle, Minas Gerais, Brazil. It is a threatened species, which is suffering from habitat loss due to the growth of cities and mining activities. The knowledge of genetic variability in plant populations is one of the main branches of conservation genetics, linking genetic data to conservation strategies while the knowledge about plant reproductive biology can aid in understanding key aspects of their life story, as well as in the comprehension of their distribution and survival strategies. Thus, the study of diversity, richness, and genetic structure, as well as the reproductive biology of populations of V. minarum can contribute to the development of conservation actions. Chapter 1 presents the transferability of 14 microsatellite loci for V. minarum. Among the results of this chapter, we highlight the successful transferability of 10 microsatellite loci described for other species of Bromeliaceae, all of which are polymorphic. In Chapter 2, we present the genetic analyses of 12 populations of V. minarum that are distributed throughout the Iron Quadrangle. We used the 10 microsatellite loci tested in Chapter 1. The results show a low population structuring (Fst = 0.088), but with different values of genetic richness (mean = 2.566) and gene diversity (mean = 0.635) for all populations; and a high inbreeding coefficient (Gis = 0.376). These may be the result of pollinators action and/or efficient seed dispersal, thus allowing a high connectivity among populations of naturally fragmented outcrops. The reproductive biology and floral morphology of a population of V. minarum, located in the Parque Estadual da Serra do Rola-Moça, are studied in Chapter 3. This reserve is the only public environmental protection area where the species occurs. As a result of field experiments and observations, we found that the species has its flowering period from January to March, with flowers that last for two days and that it has a mixed pollination syndrome. It is primarily alogamous, but also has the capacity to be self-ferilized. It is expected that data obtained in chapters 1, 2 and 3 serve as basis for other studies with species from the ferruginous rocky fields, since until now, to our knowledge, there are no other survey of endemic species from the Iron Quadrangle, seeking to merge the genetic knowledge, with the data of the reproductive biology, with the ultimate aim of biodiversity conservation. Considering the great habitat loss for the species by mining, it becomes crucial to analyze the creation of new protected areas for its conservation

Análise de uso e cobertura do solo no município de Parelhas/RN

The human interference in the semiarid region of Seridó Potiguar has promoted the increase of degraded areas. The economic dynamic that was established in the Seridó territory, especially after the fall of the trinomial cattle-cotton-mining in the 70s and 80s of the 20th century as pillars of the regional economy, resulted in an accelerated process of erosion of natural resources. The municipalities of the Seridó region have been spatially reordered by this new economic dynamic, marked by the growth of existing enterprises, and the development of new agricultural practices. One of the municipalities in the region that restructured its territorial space with the emergence of new agro-industrial activities was the town of Parelhas. With the demise of the trinomial cattle-cotton-mining in the 1980s, other productive activities were intensified from the 1990s, amongst them, pottery, responsible for the vegetal extraction for use as energy source. This recent economic and spatial restructuring in the region, reflected in the Parelhense municipal territory, required new productive ingredients responsible for the modification of past production relations that were based on cattle, cotton and mining. By that a process of exploring the environment was unleashed, especially the native vegetation, in an uncontrolled manner. In this context, the objective of this study was to survey and detect deforestation in the areas of Caatinga vegetation, used indiscriminately as energy supply for new agricultural practices, using remote sensing techniques based on the quantification of the Normalized Difference Vegetation Index / NDVI, Soil-Adjusted Vegetation Index / SAVI, surface temperature and rainfall data in the years 1990 and 2010. The results indicated that SAVI values above 0.2 in 1990 and 2010 represent the areas with the highest density of vegetation that occur exclusively along the major drainages in the town and areas of higher elevations. The areas between the ranges of values from 0.5 to 0.15 SAVI are areas with poor vegetation. On the other hand the highest values of temperature are distributed in the western and southeastern parts of the township, usually in places where the soil is exposed or there is sparse vegetation. The areas of bare soil decreased in extension in 2010 at 11, 6% when related to 1990, this was caused by a higher rainfall intensity in the first half of 2010, but no regeneration of vegetation occurred in some places in the western and southeastern areas of the municipality today, due to the extraction of firewood to fuel the furnaces of industries in town

Uso do GPR na geração de modelos estáticos 3D para o planejamento de lavra de minerais pesados depositados em dunas eólicas

This project was developed as a partnership between the Laboratory of Stratigraphical Analyses of the Geology Department of UFRN and the company Millennium Inorganic Chemicals Mineração Ltda. This company is located in the north end of the paraiban coast, in the municipal district of Mataraca. Millennium has as main prospected product, heavy minerals as ilmenita, rutilo and zircon presents in the sands of the dunes. These dunes are predominantly inactive, and overlap the superior portion of Barreiras Formation rocks. The mining happens with the use of a dredge that is emerged at an artificial lake on the dunes. This dredge removes sand dunes of the bottom lake (after it disassembles of the lake borders with water jets) and directs for the concentration plant, through piping where the minerals are then separate. The present work consisted in the acquisition external geometries of the dunes, where in the end a 3D Static Model could be set up of these sedimentary deposits with emphasis in the behavior of the structural top of Barreiras Formation rocks (inferior limit of the deposit). The knowledge of this surface is important in the phase of the plowing planning for the company, because a calculation mistake can do with that the dredge works too close of this limit, taking the risk that fragments can cause obstruction in the dredge generating a financial damage so much in the equipment repair as for the stopped days production. During the field stages (accomplished in 2006 and 2007) topographical techniques risings were used with Total Station and Geodesic GPS as well as shallow geophysical acquisitions with GPR (Ground Penetrating Radar). It was acquired almost 10,4km of topography and 10km of profiles GPR. The Geodesic GPS was used for the data geopositioning and topographical rising of a traverse line with 630m of extension in the stage of 2007. The GPR was shown a reliable method, ecologically clean, fast acquisition and with a low cost in relation to traditional methods as surveys. The main advantage of this equipment is obtain a continuous information to superior surface Barreiras Formation rocks. The static models 3D were elaborated starting from the obtained data being used two specific softwares for visualization 3D: GoCAD 2.0.8 and Datamine. The visualization 3D allows a better understanding of the Barreiras surface behavior as well as it makes possible the execution of several types of measurements, favoring like calculations and allowing that procedures used for mineral extraction is used with larger safety

Estudo geoquímico e geocronológico Rb-Sr e Sm-Nd em zonas de cisalhamento mineralizadas em ouro e suas relações com as rochas encaixantes e geocronológico Sm-Nd em mineralizações de scheelita na faixa Seridó

Great part of the gold mineralizations are associated with shearing zones through which circulate a great volume of fluids, that interact with the host rocks, originating leaching or precipitation of chemical elements, including gold. The studied mineralizations are inserted in the Seridó Belt. The tungsten mineralization in Brejuí Mine is hosted in calcsilicate rocks from Jucurutu Formation. The São Francisco auriferous mineralization has as host rocks mica-schists from Seridó Formation, while the Ponta da Serra and Fazenda Simpático mineralizations are hosted in orthogneisses of this fold belt basement. The research conducted on these mineralizations had the purpose of integrate the data of chemical elements behavior during the shearing/mineralizing event, and its influence on the isotopic systems Rb-Sr and Sm-Nd. The studies of chemical mobility in the auriferous mineralizations showed that elements that during the shearing displayed in general an immobile behavior were Al, Ti and Zr. Among the elements that were mobilized during the event, K and Rb showed mass gain in ali belts of transformed rocks, while the elements Ca, Na and Sr normally lost mass. Petrographic studies showed that the minerais biotite and plagioclase, in all investigated mineralizations, played an important role in the chemical reactions occurred in the transformed rocks to the generation of muscovite, cordierite and sillimanite, justifying the input of K to the formation of muscovite, and the release of Na and Ca from plagioclase to the fluid phase. In the São Francisco auriferous mineralization, the results of the Rb-Sr isotopic analysis yielded ages of 645 ± 19 Ma and 596 ± 17 Ma, with both samples, from original and transformed rocks. Two ages, 569 ± 20 Ma. and 554 ± 19 Ma., were obtained with samples frem the transformed rocks domain. These ages suggest that there were two metamorphic pulses during the emplacement of the mineralized shearing zone. The Sm-Nd data yielded TDM ages of 1,31 Ga and 1,26 Ga with 3Nd (0,6 Ga) of -0,26 e -0,40 for the original and final transformed rocks, respectively. In case of the orthogneisses of Caicó Complex, e.g. the Ponta da Serra and Fazenda Simpático mineralizations, the Rb-Sr data did not yield ages with geological significance. In the Ponta da Serra mineralization, the Sm-Nd isotopic data yielded T DM ages of 2,56 Ga and 2,63 Ga to the original rocks and of 2,71 Ga to the mineralized sheared rock, and values of 3Nd (2,0 Ga) between -3,70 e -5,42 to the original and sheared rock, respectively. In the Fazenda Simpático, Sm-Nd data yielded TDM between 2,65 and 2,69 Ga with values of 3Nd (2,0 Ga) between -5,25 e -5,52. Considering the Sm-Nd data, the TDM ages may be admitted as the age of the parental magma extraction, producer of the protoliths of the orthogneisses from Ponta da Serra and Fazenda Simpático mineralizations. The chemical mobility studies showed that in the basement hosted mineralizations, Rb achieved mass while Sr lost mass, as Sm as well as Nd were strongly mobilized. The Sm/Nd ratio remained constant, however, confirming the isochemical character of those elements. In the basement mineralizations, Rb-Sr ages are destituted of geological significance, because of the partial opening of the isotopic system during the tectono-metamorphic transformations. In the tungsten mineralization, the diagram Sm-Nd constructed with the whole-rock data of calcsilicatic and the high-temperature paragenesis (garnet, diopside and iron-pargasitic hornblende) indicated an 631 ± 24 Ma age, while with the whole-rock data and low-temperature paragenesis (vesuvianite, epidote and calcite), a 537 ± 107 Ma age was obtained. These ages, associated with the petrographic observations, suggest that there was a time gap among the hydrothernal events responsible by the formation of the high and low temperature paragenesis in the calcsilicatic rocks mineralized in scheelite

A Produção do espaço das pequenas cidades do Seridó Potiguar

The present research wants to comprehend the small towns process of production situated in the Seridó Potiguar, under a historical perspective. Many were the ways passedthrough and the social agents involved in the organization of this space that is build in the treefold: cattle-breeding/cotton plantation/mining, giving origin to the firstcenters. Through the analysis of demographic, economical and social aspects we elaborate a small towns urban profile, considering the 80 s period until today. Those aspects reveal a town s systems with several characteristics and impossible to be insert in some table of reference. These elements are showed in the concentration of urban equipment in the distribution of economical opportunities and in the social connection that are made in the routine of each place. However, the analysis of production s process and special organization proportinated us an explanation of small towns reality, confirming its importance in the urban system, catching sight that, although the restrict urban functions presented, it confirms its towns character by organizing andadministratingthe life specific community

Uma abordagem para evolução e reconciliação de linhas de produtos de software clonadas

Software product line engineering promotes large software reuse by developing a system family that shares a set of developed core features, and enables the selection and customization of a set of variabilities that distinguish each software product family from the others. In order to address the time-to-market, the software industry has been using the clone-and-own technique to create and manage new software products or product lines. Despite its advantages, the clone-and-own approach brings several difficulties for the evolution and reconciliation of the software product lines, especially because of the code conflicts generated by the simultaneous evolution of the original software product line, called Source, and its cloned products, called Target. This thesis proposes an approach to evolve and reconcile cloned products based on mining software repositories and code conflict analysis techniques. The approach provides support to the identification of different kinds of code conflicts – lexical, structural and semantics – that can occur during development task integration – bug correction, enhancements and new use cases – from the original evolved software product line to the cloned product line. We have also conducted an empirical study of characterization of the code conflicts produced during the evolution and merging of two large-scale web information system product lines. The results of our study demonstrate the approach potential to automatically or semi-automatically solve several existing code conflicts thus contributing to reduce the complexity and costs of the reconciliation of cloned software product lines.

Utilização do resíduo grosso do beneficiamento da scheelita em aplicações rodoviárias

Mining is an activity of great economic and social value, assisting in the development of the country. However, it can be extremely harmful to the environment if no proper waste management study exists as mitigation measure its effects. Sample some of these harmful effects are pollution: water, through the seepage of waste into the groundwater; soil; of fauna and flora; sound (due to the noise of machines); visual from the residue stored in the open, changing the local landscape; and air. One way to mitigate environmental impacts caused by mining is the proper management of their waste through their use on highways. To that end, this paper proposes to give an appropriate destination to grit coming from the beneficiation of scheelite, due to mining activity from mining group Tomaz Salustino in Brejuí mine, located in the city of Currais Novos in the state of Rio Grande do Norte. This work was developed in four stages. The first comprised the chemical and mineralogical tests, DRX and FRX in which they sought to discover the composition of the material studied. The next step involved the physical characteristics of the waste by means of specific tests and grinding the solid mass, LL and LP. The third stage included the specific tests applied to the pavement, with the compaction test and test Index Support California. Finally, the fourth stage was the mechanical characterization, represented by direct shear tests, both in the flooded condition and not in flooded condition. The technical feasibility of using the modified energie compacted in layers of subbase power has been verified. In normal and intermediate energies is feasible to use less noble as layers as the subgrade. The incorporation of the waste in layers of road pavements provide an alternative to conventionally used in paving aggregates, providing a proper disposal of tailings from scheelite, as well as environmental preservation

Contribuições de modelagem de equações estruturais na análise de dados em modelos comportamentais de destino turístico

A fragilidade brasileira quanto à competitividade turística é um fato observável nos dados da Organização Mundial do Turismo. O Brasil caiu em 2011, da 45ª para a 52ª posição, apesar de liderar no atributo recursos naturais e estar colocado na 23° em recursos culturais. Assim, grandes interesses e esforços têm sido direcionados para o estudo da competitividade dos produtos e destinos turísticos. O destino turístico é caracterizado por um conjunto complexo e articulado de fatores tangíveis e intangíveis, apresentando alta complexidade, dados de elevada dimensionalidade, não linearidade e comportamento dinâmico, tornando-se difícil a modelagem desses processos por meio de abordagens baseadas em técnicas estatísticas clássicas. Esta tese investigou modelos de equações estruturais e seus algoritmos, aplicados nesta área, analisando o ciclo completo de análise de dados, em um processo confirmatório no desenvolvimento e avaliação de um modelo holístico da satisfação do turista; na validação da estrutura do modelo de medida e do modelo estrutural, por meio de testes de invariância de múltiplos grupos; na análise comparativa dos métodos de estimação MLE, GLS e ULS para a modelagem da satisfação e na realização de segmentação de mercado no setor de destino turístico utilizando mapas auto-organizáveis de Kohonen e sua validação com modelagem de equações estruturais. Aplicações foram feitas em análises de dados no setor de turismo, principal indústria de serviços do Estado do Rio Grande do Norte, tendo sido, teoricamente desenvolvidos e testados empiricamente, modelos de equações estruturais em padrões comportamentais de destino turístico. Os resultados do estudo empírico se basearam em pesquisas com a técnica de amostragem aleatória sistemática, efetuadas em Natal-RN, entre Janeiro e Março de 2013 e forneceram evidências sustentáveis de que o modelo teórico proposto é satisfatório, com elevada capacidade explicativa e preditiva, sendo a satisfação o antecedente mais importante da lealdade no destino. Além disso, a satisfação é mediadora entre a geração da motivação da viagem e a lealdade do destino e que os turistas buscam primeiro à satisfação com a qualidade dos serviços de turismo e, posteriormente, com os aspectos que influenciam a lealdade. Contribuições acadêmicas e gerenciais são mostradas e sugestões de estudo são dadas para trabalhos futuros.

Metodologia para implantação de um sistema de disposição de rejeitos em minério de ferro na região do semiárido: estudo de caso

Tailings dams are structures that aims to retain the solid waste and water from mining processes. Its analysis and planning begins with searching of location for deployment, step on which to bind all kinds of variables that directly or indirectly influence the work, such as geological, hydrological, tectonic, topographic, geotechnical, environmental, social characteristics, evaluation security risks, among others. Thus, this paper aims to present a study on the most appropriate and secure type of busbar to design a layout structure of iron ore tailings, taking into account all the above mentioned variables. The case study involves the assessment of sites for location of dams of tailings disposal beneficiation of iron mine to be built in Bonito, in the municipality of Jucurutu in Seridó Potiguar. For site selection among alternatives, various aspects of the current state of the art were considered, one that causes the least environmental impact, low cost investment, adding value to the product and especially the safety of the implanted structure mitigates the concern about induced earthquakes as a result of liquefaction wastes somatized by dams in the region, as the tilling of Mina Bonito is located practically in the hydraulic basin dam Armando Ribeiro in environmental protection (APA). The methodology compares induced by dams in the semiarid region with the characteristics of the waste disposal and sterile seismicity, taking into account the enhancement of liquefaction by the action of seismicity in the Mina Bonito region. With the fulcrum in the methodology, we indicated the best busbar type for disposal of tailings from iron ore or combination of them, to be designed and built in semiarid particularly for Mina Bonito. Also presents a number of possible uses for the tailings and in engineering activities, which may cause processing to the common good.

Caracterização de resíduos oriundos do beneficiamento do minério de ferro para uso em pavimentação

Mining in Brazil has a key role in economic and social development, contributing directly to improve the lives of the population. However, the mining activity even if done responsibly and with a proper study of waste management to reduce the impact of its effects, may cause harmful damage to the environment. Other forms of pollution are also caused due to mining activity. The visual pollution caused by the waste storage at open sky, in addition to the noise pollution caused by the excessive noise of the machines both in the extraction of ore, as in processing. An alternative way to lessen the environmental impacts caused by mining is the use of waste in layers that will compose the pavements along the highways. Thus, this work sets out to give a proper disposal of the wastes from the processing of iron ore, resulting from the mining activity of the group of mining Mhag Services and Mining S/A, in the mine of Bonito, located in Jucurutu, a town in the state of Rio Grande do Norte. The residues of the iron ore were stabilized with a granular soil from the city of Macaiba, also in the state of Rio Grande do Norte, which is being used in the duplication of the BR-304 referring to the entitled passage of Reta Tabajara. The present work was developed in three stages, being the first one divided by the chemical and mineralogical tests, by the tests of physical characterization and by the tests of paving for the residues of the iron ore. The second stage corresponds to the same tests being performed for granular soil. The third stage includes the essays abovementioned for three different mixtures of iron ore waste and granular soil, being they: 15% of iron-ore waste and 85% of granular soil, 25% of iron-ore waste and 75% of granular soil, 50% of iron-ore waste and 50% of granular soil. The technical feasibility of using waste from the iron ore beneficiation was checked, compressed in the intermediate energy and modified for use in base layers, sub-base, reinforcement subgrade and subgrade. The incorporation of the residues originating from the improvement of the iron ore in highways will provide an alternative to the use of aggregate conventionally used in the paving, besides preserving the environment.

Utilizando comitês de classificadores para predição de rendimento escolar

Educational Data Mining is an application domain in artificial intelligence area that has been extensively explored nowadays. Technological advances and in particular, the increasing use of virtual learning environments have allowed the generation of considerable amounts of data to be investigated. Among the activities to be treated in this context exists the prediction of school performance of the students, which can be accomplished through the use of machine learning techniques. Such techniques may be used for student’s classification in predefined labels. One of the strategies to apply these techniques consists in their combination to design multi-classifier systems, which efficiency can be proven by results achieved in other studies conducted in several areas, such as medicine, commerce and biometrics. The data used in the experiments were obtained from the interactions between students in one of the most used virtual learning environments called Moodle. In this context, this paper presents the results of several experiments that include the use of specific multi-classifier systems systems, called ensembles, aiming to reach better results in school performance prediction that is, searching for highest accuracy percentage in the student’s classification. Therefore, this paper presents a significant exploration of educational data and it shows analyzes of relevant results about these experiments.