Silhouette-based clustering using an immune network
Tipo
Artigo de evento
Data de publicação
2012
Periódico
2012 IEEE Congress on Evolutionary Computation, CEC 2012
Citações (Scopus)
6
Autores
Borges E.
Ferrari D.G.
De Castro L.N.
Ferrari D.G.
De Castro L.N.
Orientador
Título da Revista
ISSN da Revista
Título de Volume
Membros da banca
Programa
Resumo
Clustering is an important Data Mining task from the field of Knowledge Discovery in Databases. Many algorithms can perform clustering in a simple and efficient manner, but have drawbacks, such as the lack of a way to automatically determine the optimal number of clusters in the dataset and the possibility of getting stuck in local optima solutions. To try and reduce these drawbacks this work proposes a new clustering algorithm based on Artificial Immune Systems. This algorithm is characterized by the generation of multiple simultaneous high quality solutions in terms of the number of clusters in the database and the use of a cost function that explicitly evaluates the quality of clusters, minimizing the inconvenience of getting stuck in local optima solutions. © 2012 IEEE.
Descrição
Palavras-chave
Assuntos Scopus
Artificial Immune System , Clustering , Data mining tasks , Data sets , Diversity , High-quality solutions , Immune network , K-means , Knowledge discovery in database , Local optima , Number of clusters , Optimal number