Agrupamento e classificação de dados utilizando um algoritmo inspirado no comportamento de abelhas

• Autor(a) principal: Cruz, Dávila Patrícia Ferreira
• Data de Publicação: 2015
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações do Mackenzie
• Texto Completo: http://dspace.mackenzie.br/handle/10899/24377
• topic: agrupamento; classificação; inteligência de enxame; algoritmo bioinspirado; clustering; classification; swarm intelligence; bioinspired algorithms; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA

With the popularization of Internet, the advancement of electronic devices and the ease of storage, the volume of data stored and available at companies has increased substantially. Therefore, it becomes necessary to use intelligent techniques to extract useful information and knowledge from these data. In this context, Data Mining has been the aim of several researches by providing a set of intelligent techniques to the exploration of large volumes of data. The present project aims to research and develop new algorithms inspired by the collective behavior of bee colonies for solving complex clustering and classification tasks. More specifically, this project proposes adaptations of an optimization algorithm inspired by the behavior of bees so that it can be applied to solve clustering problems and also for positioning centers of RBF neural networks. The proposed approaches were applied to several benchmark problems with promising results.