Title :
A Fuzzy Variant of an Evolutionary Algorithm for Clustering
Author :
Alves, Vinícius S. ; Campello, Ricardo J G B ; Hruschka, Eduardo R.
Author_Institution :
Catholic Univ. of Santos, Santos
Abstract :
A fuzzy version of an evolutionary algorithm for clustering (EAC) proposed in previous work is introduced. This algorithm uses a fuzzy cluster validity criterion and a fuzzy local search algorithm instead of their hard counterparts employed by EAC. It is shown by means of theoretical complexity analyses that this algorithm can be more efficient than systematic (i.e. repetitive) approaches when the number of clusters is unknown. An illustrative example with computational experiments and statistical analyses is also presented.
Keywords :
evolutionary computation; pattern clustering; search problems; evolutionary algorithm; fuzzy cluster validity criterion; fuzzy local search algorithm; Algorithm design and analysis; Clustering algorithms; Computer science; Evolutionary computation; NP-hard problem; Partitioning algorithms; Statistical analysis;
Conference_Titel :
Fuzzy Systems Conference, 2007. FUZZ-IEEE 2007. IEEE International
Conference_Location :
London
Print_ISBN :
1-4244-1209-9
Electronic_ISBN :
1098-7584
DOI :
10.1109/FUZZY.2007.4295395
