Research Article

An Island Grouping Genetic Algorithm for Fuzzy Partitioning Problems

Table 4

Comparison of the results (in terms of the number of clusters finally found and as a function of the Rand index) obtained by the proposed GGA algorithm with MDB index and different distances and the FCM algorithm (with Euclidean distance, which has been found to be the best for this algorithm) in the third synthetic clustering problem considered. See the main text for further details.

Algorithm Distance Number of clusters Rand index

GGA (MDB index) Euclidean 5 0.8989
GGA (MDB index) GK 6 0.9475
GGA (MDB index) AFC 6 0.9670
FCM Euclidean 6 0.9416