• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 2
  • Tagged with
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

O problema de clustering heterog?neo fuzzy: modelos e heur?sticas

Pinheiro, Daniel Nobre 27 January 2017 (has links)
Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2017-04-03T19:47:15Z No. of bitstreams: 1 DanielNobrePinheiro_DISSERT.pdf: 900596 bytes, checksum: 82c38f5d0fc71d5fb71fb0c1acd283c6 (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2017-04-06T19:26:01Z (GMT) No. of bitstreams: 1 DanielNobrePinheiro_DISSERT.pdf: 900596 bytes, checksum: 82c38f5d0fc71d5fb71fb0c1acd283c6 (MD5) / Made available in DSpace on 2017-04-06T19:26:01Z (GMT). No. of bitstreams: 1 DanielNobrePinheiro_DISSERT.pdf: 900596 bytes, checksum: 82c38f5d0fc71d5fb71fb0c1acd283c6 (MD5) Previous issue date: 2017-01-27 / Este trabalho prop?e formula??es para o Problema de Clustering Heterog?neo Fuzzy, assim como um m?todo heur?stico de Busca em Vizinhan?a Vari?vel para resolv?-lo. O Problema de Clustering Heterog?neo Fuzzy ? um problema de agrupamento de dados modelado em dois n?veis. O primeiro identifica grupos de indiv?duos cujas percep??es acerca dos objetos envolvidos sejam similares. O segundo n?vel identifica parti??es fuzzy de objetos para cada grupo de indiv?duos. O segundo n?vel ? baseado no problema das p-medianas, cujo objetivo ? particionar um conjunto de objetos em subconjuntos menores e definir um objeto para cada subconjunto como mediana, de modo que a soma das dissimilaridades entre cada objeto e sua mediana seja m?nima. O Problema de Clustering Heterog?neo Fuzzy generaliza o problema das p-medianas para ambientes fuzzy, permitindo que os n?veis de pertin?ncia de cada objeto em rela??o a cada cluster sejam fracion?rios. Essa generaliza??o permite novas interpreta??es dos resultados, como a identifica??o de rela??es simult?neas de objetos com diferentes clusters. / This work proposes formulations for the Fuzzy Heterogeneous Clustering Problem, as well as a heuristic method of Variable Neighborhood Search to solve it. The Fuzzy Heterogeneous Clustering Problem is a clustering problem that is formulated in two levels. The first identifies groups of individuals whose perceptions about the objects involved are similar. The second level identifies fuzzy partitions of objects for each group of individuals. The second level is based on the p-median problem, whose objective is to partition a set of objects into smaller subsets and to define an object as median for each subset, such that the sum of dissimilarities between each object and its median is minimal. The Fuzzy Heterogeneous Clustering Problem generalizes the p-median problem to fuzzy environments, allowing the degrees of membership between each object and each cluster to be fractionary. This generalization allows new interpretations about the results, such as the identification of simultaneous relationships of objects with different clusters.
2

Formula??es e algoritmos para o problema das p-medianas heterog?neo livre de penalidade

Santi, ?verton 14 November 2014 (has links)
Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2016-01-05T18:01:11Z No. of bitstreams: 1 EvertonSanti_TESE.pdf: 601652 bytes, checksum: 52767a19768856b40fcce8bb5611ef4b (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2016-01-11T18:20:39Z (GMT) No. of bitstreams: 1 EvertonSanti_TESE.pdf: 601652 bytes, checksum: 52767a19768856b40fcce8bb5611ef4b (MD5) / Made available in DSpace on 2016-01-11T18:20:39Z (GMT). No. of bitstreams: 1 EvertonSanti_TESE.pdf: 601652 bytes, checksum: 52767a19768856b40fcce8bb5611ef4b (MD5) Previous issue date: 2014-11-14 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior - CAPES / Apresenta-se neste trabalho um novo modelo para o Problema das p-Medianas Heterog?neo (PPMH), proposto para recuperar a estrutura de categorias n?o-observadas presente em dados oriundos de uma tarefa de triagem, uma abordagem popular que possibilita entender a percep??o heterog?nea que um grupo de indiv?duos tem em rela??o a um conjunto de produtos ou marcas. Este novo modelo ? chamado Problema das p-Medianas Heterog?neo Livre de Penalidade (PPMHLP), uma vers?o mono-objetivo do problema original, o PPMH. O par?metro principal do modelo PPMH ? tamb?m eliminado, o fator de penalidade. Este par?metro ? respons?vel pela pondera??o dos termos de sua fun??o objetivo. O ajuste do fator de penalidade controla a maneira como o modelo recupera a estrutura de categorias n?o-observadas presente nos dados e depende de um amplo conhecimento do problema. Adicionalmente, duas formula??es complementares para o PPMHLP s?o apresentadas, ambas problemas de programa??o linear inteira mista. A partir destas formula??es adicionais, limitantes inferiores foram obtidos para o PPMHLP. Estes valores foram utilizados para validar um algoritmo de Busca em Vizinhan?a Variada (VNS), proposto para resolver o PPMHLP. Este algoritmo obteve solu??es de boa qualidade para o PPMHLP, resolvendo inst?ncias geradas de forma artificial por meio de uma Simula??o de Monte Carlo e inst?ncias reais, mesmo com recursos computacionais limitados. As estat?sticas analisadas neste trabalho sugerem que o novo algoritmo e modelo, o PPMHLP, pode recuperar de forma mais precisa que o algoritmo e modelo original, o PPMH, a estrutura de categorias n?o-observadas presente nos dados, relacionada ? percep??o heterog?nea dos indiv?duos. Por fim, uma exemplo de aplica??o do PPMHLP ? apresentado, bem como s?o consideradas novas possibilidades para este modelo, estendendo-o a ambientes fuzzy / This work presents a new model for the Heterogeneous p-median Problem (HPM), proposed to recover the hidden category structures present in the data provided by a sorting task procedure, a popular approach to understand heterogeneous individual?s perception of products and brands. This new model is named as the Penalty-free Heterogeneous p-median Problem (PFHPM), a single-objective version of the original problem, the HPM. The main parameter in the HPM is also eliminated, the penalty factor. It is responsible for the weighting of the objective function terms. The adjusting of this parameter controls the way that the model recovers the hidden category structures present in data, and depends on a broad knowledge of the problem. Additionally, two complementary formulations for the PFHPM are shown, both mixed integer linear programming problems. From these additional formulations lower-bounds were obtained for the PFHPM. These values were used to validate a specialized Variable Neighborhood Search (VNS) algorithm, proposed to solve the PFHPM. This algorithm provided good quality solutions for the PFHPM, solving artificial generated instances from a Monte Carlo Simulation and real data instances, even with limited computational resources. Statistical analyses presented in this work suggest that the new algorithm and model, the PFHPM, can recover more accurately the original category structures related to heterogeneous individual?s perceptions than the original model and algorithm, the HPM. Finally, an illustrative application of the PFHPM is presented, as well as some insights about some new possibilities for it, extending the new model to fuzzy environments

Page generated in 0.1499 seconds