Agrupamento baseado em SOM com pesos adaptativos para múltiplas tabelas de dissimilaridade

Anderson B. S. DantasFrancisco de A. T. de Carvalho

This paper introduces a clustering algorithm based on batch Self- organizing map to partition objects taking into account their relational descriptions given by multiple dissimilarity matrices. The presented approach provide a partition of the objects and a prototype for each cluster, moreover the method is capable of learn relevance weights for each dissimilarity matrix by optimizing an adequacy criterion that measures the fit between clusters and the respectives prototypes. These relevance weights change at each iteration and are differ- ent from one cluster to another. Experiments using real-world data bases are considered to show the usefulness of the method.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: