Clusterização através de árvores de padrões fuzzy e programação genética cartesiana

• Autor(a) principal: Araujo, Vivian de Oliveira
• Data de Publicação: 2017
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UERJ
• Texto Completo: http://www.bdtd.uerj.br/handle/1/11867
• topic: Electronic Engineering; Grouping of related data; Diffuse logic; Machine Learning; Clustering; Cartesian Genetic Programming; Clustering; Interpretability; Engenharia eletrônica; Agrupamento de dados relacionados; Lógica difusa; Aprendizado de máquina; Árvores Fuzzy de Padrões; Programação Genética Cartesiana; Clusterização; Interpretabilidade; CNPQ::ENGENHARIAS

This work presents o method for fuzzy clustering. Instead of the traditional fuzzy based rules, it was used a model called Fuzzy Pattern Trees (FPT), which is a hierarchical tree-based model, having as internal nodes, fuzzy logical operators and the leaves are composed of a combination of fuzzy terms with the input attributes. The method was obtained by creating a tree for each cluster, this tree will be a logic class description which allows the interpretation of the results. The learning method originally designed for FPT was replaced by Cartesian Genetic Programming where the fitness function reflects the quality of the clustering obtained through different indices. The FPT method was compared against other clustering techniques, such as: k-means, k-medoids, Agglomerative, Fuzzy C-means, Kohonen and DBSCAN on several datasets from artificial bases and the UCI Machine Learning Repository and it presented competitive results. It was also used to solve a segmentation problem from a mobile operator with promising results.