Hi-index | 0.00 |
The well known fuzzy partition clustering algorithms are most based on Euclidean distance function, which can only be used to detect spherical structural clusters. Many improved algorithms have been developed to detect non-spherical structural clusters. In our previous work, vectors were represented as objects of the feature space, we found that the difference of vectors can reflect the shape similarity message of these different objects, and we proposed the morphology similarity distance (MSD) for similarity estimation. In this paper, an improved fuzzy partition clustering algorithm, “fuzzy c-mean based on the morphology similarity distance (FCM-MSD)”, is proposed. This new algorithm has been tested on the Iris data set from the UCI repository. Experiment results prove that the performance of the FCM-MSD algorithm is better than those of the FCM algorithm based on the traditional Euclidean and Manhattan distances.