▼ 

Submitted by Marcos Samuel ([email protected]) on 2017-06-13T13:35:56Z No. of bitstreams: 1 Disserta????o Ligia.pdf: 1370436 bytes, checksum: e38350a980574846d81de84112d043d4 (MD5)

Approved for entry into archive by Vanessa Reis ([email protected]) on 2017-06-27T12:45:35Z (GMT) No. of bitstreams: 1 Disserta????o Ligia.pdf: 1370436 bytes, checksum: e38350a980574846d81de84112d043d4 (MD5)

Made available in DSpace on 2017-06-27T12:45:35Z (GMT). No. of bitstreams: 1 Disserta????o Ligia.pdf: 1370436 bytes, checksum: e38350a980574846d81de84112d043d4 (MD5)

Partindo da premissa de que os problemas s??o os instrumentos pelos quais a Matem??tica sempre se desenvolveu e que resolv??-los ?? o que impulsiona n??o s?? o seu ensino, mas o desenvolvimento desta Ci??ncia, o presente trabalho aborda a tem??tica dos problemas matem??ticos sob perspectiva da elabora????o dos mesmos. Este material apresenta um resumo te??rico sobre alguns aspectos da resolu????o de problemas onde buscou-se atrav??s do m??todo heur??stico organizado por George Polya e de estudos derivados da obra do autor expor alguns temas tais como a escrita matem??tica, os tipos de problemas, as etapas e estrat??gias de resolu????o e os elementos que caracterizam um bom problema matem??tico. T??picos fundamentais que alicer??aram e instrumentalizaram uma oficina com estrat??gias para a elabora????o de problemas matem??ticos voltados para o Ensino M??dio, o objeto central deste estudo, juntamente com o relato de sua aplica????o em uma turma de licenciandos em Matem??tica.

Advisors/Committee Members: Mello, Vinicius Moreira, Mello, Vinicius Moreira, Feitosa, Samuel Barbosa, Silva, Rita C??ssia de Jesus.

▼ 

Submitted by Setor de Tratamento da Informa??o - BC/PUCRS ([email protected]) on 2016-07-26T14:20:24Z No. of bitstreams: 1 DIS_RAMON_FRAGA_PEREIRA_COMPLETO.pdf: 1672915 bytes, checksum: 3cf47a5c224b3f99a343aab57bf00bf3 (MD5)

Made available in DSpace on 2016-07-26T14:20:24Z (GMT). No. of bitstreams: 1 DIS_RAMON_FRAGA_PEREIRA_COMPLETO.pdf: 1672915 bytes, checksum: 3cf47a5c224b3f99a343aab57bf00bf3 (MD5) Previous issue date: 2016-03-15

Recognition of goals and plans using incomplete evidence from action execution can be done efficiently by using automated planning techniques. In many applications it is important to recognize goals and plans not only accurately, but also quickly. In order to address this challenge, we develop recognition approaches based on planning techniques that rely on planning landmarks to filter candidate goals and plans from observations. In automated planning, landmarks are properties or actions that cannot be avoided to achieve a goal. We address the task of recognizing goals and plans without pre-defined static plan libraries, and instead we use a planning domain definition to represent the problem and the expected agent behavior. In this work, we show the applicability of planning techniques for recognition tasks in three settings: first, we use planning landmarks to develop a heuristic-based plan recognition approach; second, we refine an existing planningbased plan recognition approach; and finally, we use planning techniques to develop an approach for detecting plan abandonment. The plan abandonment detection approach we develop aims to analyze a sequence of observations and a monitored goal to determine if an observed agent is still pursuing, or has no intention to complete such monitored goal. These recognition approaches are evaluated in experiments over several planning domains. We show that our plan recognition approach yields not only accuracy comparable to other state-of-the-art techniques, but also substantially lower recognition time over such techniques. Furthermore, our plan abandonment detection approach yields high accuracy at low computational cost to detect which actions do not contribute for achieving a particular monitored goal.

T?cnicas de planejamento autom?tico s?o eficientes no reconhecimento de objetivos e planos a partir da execu??o de a??es e evid?ncias incompletas. Para muitas aplica??es ? importante reconhecer objetivos e planos n?o somente acuradamente, mas tamb?m de maneira r?pida e precisa. Assim, para lidar com esse desafio, desenvolvemos uma abordagem a qual utiliza uma heur?stica baseada em t?cnicas de planejamento autom?tico, guiando-se por pontos-de-refer?ncia, que filtra poss?veis objetivos e planos a partir de observa??es. Em planejamento autom?tico, pontosde- refer?ncia s?o propriedades (ou a??es), em que todo o plano precisa alcan?ar (ou executar), em alguma determinada parte da execu??o do plano a fim de atingir um objetivo estipulado. Neste trabalho, formalizamos a tarefa de reconhecimento de objetivos e planos sem a utiliza??o de…

Advisors/Committee Members: Meneguzzi, Felipe Rech.

▼ 

Submitted by Setor de Tratamento da Informa??o - BC/PUCRS ([email protected]) on 2016-08-02T16:30:51Z No. of bitstreams: 1 DIS_FERNANDA_TRINDADE_DEYL_COMPLETO.pdf: 2355705 bytes, checksum: a66323937eda76c731252aaddb2b2ee7 (MD5)

Made available in DSpace on 2016-08-02T16:30:51Z (GMT). No. of bitstreams: 1 DIS_FERNANDA_TRINDADE_DEYL_COMPLETO.pdf: 2355705 bytes, checksum: a66323937eda76c731252aaddb2b2ee7 (MD5) Previous issue date: 2016-03-29

Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior - CAPES

Retailing is a complex area of study, demanding the management of many phenomena, such as prices policies. Although essential to the management of any business, the academic literature regarding price strategies in the national context is scarce and not explored extensively in Brazil, nor by academy, nor by the market. Even more incipient is the Brazilian literature regarding strategies for communication and price presentation in the retail context with few studies exploring matters relative to the design of price presentation. It is also verified that there is a gap in the literature regarding the study of price typography in Marketing, specially related to the use of fonts as heuristic cues by the consumer to perceive, interpret and create judgments regarding a product or a brand. The present study, therefore, intends to answer to the following question: does the font used to present price information of a product affect the perception of price fairness by the consumer? In order to do that, this dissertation report will initially present a review of the academic literature regarding heuristic strategies for information and message processing related to the price charged by a product under the consumer?s point of view. Questions regarding the conceptualization of price fairness, price framing and typography are also reviewed in order to develop a theoretical model based in three hypothesis, all of which were tested empirically. The presence of moderation effect of the different levels of involvement, as well as of the manipulation of reference price on the relationship between font and price fairness, were later investigated. The method used for data collection was, mainly, a experimental study, in which an experiment was developed to test the hypothesis. Results indicate that, even though no direct relationship is identified between the font used in the price presentation and perception of price fairness, in situations in which the consumer has a reference price available for the product, the relationship becomes significant. However, there is no indication that the different levels of involvement affect the same relationship. In conclusion, the present research is justified, mainly, for being innovative by shifting the focus to the study of the price typography phenomenon, a subject not often explored by the Brazilian academy, and also for building upon the international literature about the use of typography as an element of price framing by analyzing its relationship with price fairness, an unprecedented…

Advisors/Committee Members: Espartel, L?lis Balestrin.

▼ The premise of unbounded rationality advocated by the Efficient Market Hypothesis is challenged by the theoretical framework that involves the Behavioral Finance, whose basis, the Prospect Theory by Kahneman and Tversky (1979), questions the Expected Utility Theory, an important element of Neoclassical Economics as basis for decision making. This research aims to replicate the empirical investigation of the seminal article by Kahneman and Tversky (1979) to evaluate the decision-making process of employees (potential investors) of a major national financial institution. The results of this study were compared to those obtained in the original article and the studies led by C??rtes (2008), Cruz, Kimura and Krauter (2003), Rogers et al. (2007), Rogers, Favato and Securato (2008), and Torralvo (2010). The questionnaire adopted was an adaptation of the originally used, so that one could test, in studied sample, the applicability of Prospect Theory, more specifically with regard to Certain, Reflection and Isolation Effects. I also analyzed the differences in the decision making process considering the attributes of the respondents (gender, age, education, occupation, income and financial dependents). The results confirmed the existence of behavioral effects and proved that a large portion of the sample presented significant inconsistency in their choices according to the principles of the Expected Utility Theory, pointing that their decisions were not made under strictly rational behavior. Furthermore, in relation to violations observed, it was not possible to present significant conclusions in concern to the attributes by using regression analysis, suggesting the need of further studies Advisors/Committee Members: Yoshinaga, Claudia Emiko (advisor), CPF:29328076846 (advisor), http://lattes.cnpq.br/9991689534310081 (advisor).

▼ 

Submitted by Caroline Xavier ([email protected]) on 2017-10-20T10:56:20Z No. of bitstreams: 1 DIS_VANESSA_STANGHERLIN_MACHADO_COMPLETO.pdf: 9505213 bytes, checksum: c355721326b3d2f7c04de64104755fac (MD5)

Made available in DSpace on 2017-10-20T10:56:20Z (GMT). No. of bitstreams: 1 DIS_VANESSA_STANGHERLIN_MACHADO_COMPLETO.pdf: 9505213 bytes, checksum: c355721326b3d2f7c04de64104755fac (MD5) Previous issue date: 2015-08-26

Conselho Nacional de Pesquisa e Desenvolvimento Cient?fico e Tecnol?gico - CNPq

The prediction of protein tertiary structure is a problem of Structural Bioinformatics still unsolved by science. The challenge is to understand the relationship between the amino acid sequence of a protein and a three dimensional structure, which is related to the function of these macromolecules. Among the methods related to protein structure prediction is CREF (Central Residue Fragment-based Method) proposed by Dorn & Norbert Souza (2008) for prediction of proteins? or polypeptide?s approximate 3D structure. In this thesis we present the wCReF, the Web interface for the CREF method developed with a focus on usability. With this tool the users can enter the amino acid sequence of its target protein, and get as a result the approximate 3D structure of a protein in an automated manner without the need to install all the necessary tools for their use. To define the requirements for its development were conducted usability evaluations, guided by experts on both Human-Computer Interaction and bioinformatics domain areas, in three protein structures prediction servers - I-TASSER, QUARK and Robetta - all participants at CASP (Critical Assessment of Protein Structure Prediction) competition. The inspections were conducted through the Heuristic Evaluation method using Nielsens? 10 heuristics. As a result, violations were found in all heuristics resulting in 89 usability problems. They were classified into 5 severities, 29 scored as being of high priority solution and 25 as problems to be solved immediately. Assessment results serve as guiding orientation of the key features that wCReF must have compiled in a document software requirements for its implementation. From this step was carried out prototyping and glimpsing the detection of new usability problems with end users by adapting the Ssemugabi satisfaction questionnaire. As a final product we present the wCReF server, protein structure prediction server rooted in concern about the usability and interaction with its users. Furthermore, this study can contribute to improvement of usability of existing bioinformatics applications, the prediction servers analyzed and the development of new scientific tools.

A predi??o da estrutura terci?ria de prote?nas ? um problema da Bioinform?tica Estrutural ainda n?o solucionado pela ci?ncia. O desafio ? entender a rela??o entre a sequ?ncia de amino?cidos de uma prote?na e sua estrutura tridimensional 3D, que est? relacionada ? fun??o destas macromol?culas. Dentre os m?todos relacionados ?…

Advisors/Committee Members: Souza, Osmar Norberto de.

▼ 

Made available in DSpace on 2015-04-14T14:49:13Z (GMT). No. of bitstreams: 1 418219.pdf: 4033858 bytes, checksum: 2661cfd90512a0fe826582c0489e6ce1 (MD5) Previous issue date: 2009-03-10

MPSoCs s?o sistemas multiprocessados integrados na forma de um SoC. Eles s?o tend?ncia no projeto de circuitos VLSI, pois minimizam a crise de produtividade de projeto, representada pelo descompasso entre a capacidade da tecnologia do sil?cio e a capacidade atual de projeto de SoCs. Cita-se como exemplo de MPSoCs os propostos pela Intel e pela Tilera, compostos respectivamente por 80 e 64 n?cleos de processa-mento. MPSoCs podem empregar NoCs para integrar diversos processadores, mem?rias, bem como n?cleos de hardware espec?ficos. O uso de NoCs deve-se a suas vantagens em rela??o a barramentos, entre as quais maior escalabilidade e paralelismo na comunica??o. A arquitetura alvo do presente trabalho consiste em um MPSoC heterog?neo, com utiliza??o de NoC como meio interconex?o entre os elementos de processamento, suportando a execu??o de tarefas de hardware via l?gica reconfigur?vel, e a execu??o de tarefas de software via processadores. Um dos processadores da arquitetura alvo, denominado processador gerente, ? respons?vel por: ger?ncia da ocupa??o dos recursos do sistema, escalonamento, mapeamento, e configura??o de tarefas. O mapeamento de tarefas define a posi??o de uma dada tarefa no sistema. A maioria dos trabalhos encontrados na literatura prop?e t?cnicas de mapeamento est?tico, definido em tempo de projeto, no qual todas as tarefas de uma dada aplica??o s?o mapeadas simultaneamente. Este mapeamento est?tico n?o ? adequado para cen?rios com carga din?mica de tarefas. Dado que aplica??es executando em um MPSoC podem possuir um n?mero vari?vel de tarefas, e que tal n?mero pode exceder os recursos dispon?veis, ? necess?rio realizar o mapeamento de tarefas em tempo de execu??o, mapeamento este denominado de mapeamento din?mico. O presente trabalho investiga o desempenho de heur?sticas para mapeamento din?mico de tarefas, com o objetivo de minimizar congestionamentos em NoCs. As tarefas s?o mapeadas sob demanda, de acordo com as requisi??es de comunica??o e com a ocupa??o dos canais da NoC. Os algoritmos implementados aplicam estrat?gias gulosas, onde as tarefas s?o mapeadas uma por vez. Para isso, a decis?o ? baseada na informa??o local da aplica??o, apenas relacionada ? tarefa requisitada. O algoritmo utilizado como refer?ncia nos experimentos mapeia uma dada tarefa no primeiro recurso livre encontrado. Quatro heur?sticas congestion-aware s?o propostas. Atrav?s de experimentos realizados com base na modelagem do sistema no n?vel RTL, pode-se observar redu??o de 31% na carga nos canais da NoC, de 15% na lat?ncia m?dia, e de at? 87% no n?vel m?dio de congestionamento. Tais resultados demonstram a efici?ncia das heur?sticas propostas.

Advisors/Committee Members: Moraes, Fernando Gehm.

▼ 

Made available in DSpace on 2015-04-14T14:49:58Z (GMT). No. of bitstreams: 1 445541.pdf: 1627180 bytes, checksum: 184a4b2054d700b987b074f0fcdf3122 (MD5) Previous issue date: 2011-03-15

The definitory context is part of a text or utterance that provides information about a concept based on its use. The extraction of definitions from texts is an important task in various applications. Several papers present this feature as an aid in the construction of ontologies, in the development of material for aid in translation, in question answering systems, in creation of glossaries, dictionaries, among others. Thus, this study proposes set of heuristics to make the automatic extraction of potentially definitory contexts contained in Portuguese texts. The results of these heuristics were evaluated by terminologists and obtained 35.1 % F-measure when applied in a General Geology corpus and 51.7 % F-measure when applied in a General Chemistry corpus, reducing in both more than 50%of the amount of contexts to be examined by a specialist compared with the contexts extraction through a concordancer

O contexto definit?rio ? a parte de um texto ou de um enunciado que fornece informa??o sobre um conceito, com base em seu uso. A extra??o de contextos definit?rios a partir de textos ? uma tarefa importante em v?rias aplica??es. Diversos trabalhos usam este recurso como aux?lio na constru??o de ontologias, no desenvolvimento de material de aux?lio ? tradu??o, na elabora??o de sistemas de perguntas e respostas, na cria??o de gloss?rios, dicion?rios, entre outros. Nesse sentido, este trabalho prop?e um conjunto de heur?sticas para fazer a extra??o autom?tica de contextos potencialmente definit?rios em textos de l?ngua portuguesa. Os resultados dessas heur?sticas foram avaliados por termin?logos. Os resultados mostram 35,1% de F-measure quando o conjunto de heuristicas foi aplicado a um corpus de Geologia Geral e 51,7% de F-measure quando aplicado a um corpus de Qu?mica Geral. Isso proporcionou uma redu??o, em ambos os corpus, de mais de 50% da quantidade de contextos para serem analisados pelo especialista, comparando-se com a extra??o de contextos em um concordanciador

Advisors/Committee Members: Vieira, Renata.

▼ 

153 p.

Submitted by Santiago Fabio ([email protected]) on 2012-12-17T20:13:14Z No. of bitstreams: 1 888888888.pdf: 1167614 bytes, checksum: ba56025841f10ed18a90d2d6ce7efd9b (MD5)

Made available in DSpace on 2012-12-17T20:13:15Z (GMT). No. of bitstreams: 1 888888888.pdf: 1167614 bytes, checksum: ba56025841f10ed18a90d2d6ce7efd9b (MD5) Previous issue date: 2011

O objetivo desta pesquisa foi investigar de que forma aspectos afetivos se manifestam nos julgamentos de risco e benef??cio nos investidores do mercado de a????es brasileiro por meio da heur??stica afetiva. O processo decis??rio que conduz aos julgamentos ?? complexo e multifacetado, e neste trabalho pressup??e-se que o afeto assume posi????o privilegiada nos julgamentos. Para alcan??ar o objetivo pretendido criaram-se quatro est??mulos em rela????o ao mercado de a????es: alto risco; baixo risco; alto benef??cio; baixo benef??cio. A aplica????o da pesquisa ocorreu em tr??s fases, sendo uma fase de pr??teste, uma fase de teste (Fase 1- Teste) e uma terceira fase que se constituiu na aplica????o da pesquisa propriamente dita (Fase 2 ??? Pesquisa). Na fase de pr??-teste ajustou-se o instrumento de coleta de dados, e na Fase 1 ??? Teste foi feita uma pesquisa pr??via com servidores da SEFAZ-BA. Essa amostra foi composta por 134 indiv??duos segregados aleatoriamente em quatro grupos (um grupo para cada est??mulo). Os resultados obtidos nessa fase permitiram passar para a fase seguinte. Na Fase 2 ??? Pesquisa aplicou-se o experimento em uma amostra composta por 143 investidores da bolsa de valores que operam por meio do home broker. A manifesta????o da heur??stica afetiva ocorreu no Grupo 2 por meio da diferen??a significativa de percep????o de risco e de benef??cio antes e depois do est??mulo. Para tanto, utilizou-se o teste t para m??dias. Em seguida, buscou-se nas vari??veis idade, g??nero e grau autopercebido de conhecimento do mercado de a????es as determinantes da manifesta????o da heur??stica afetiva. Por meio da regress??o log??stica m??ltipla identificouse que apenas a vari??vel grau autopercebido de conhecimento do mercado de a????es exerceu influ??ncia significativa na manifesta????o da heur??stica afetiva. Os resultados encontrados na pesquisa sugerem que a heur??stica afetiva manifesta-se mais em indiv??duos com menor grau de conhecimento, e que os fatores idade e g??nero n??o exercem influ??ncia significativa.

Salvador

Advisors/Committee Members: Bruni, Adriano Leal.