Loading...
Please wait, while we are loading the content...
Similar Documents
OF DATA SCIENCE AND ITS APPLICATIONS Ensemble Based Gustafson Kessel Fuzzy Clustering
| Content Provider | Semantic Scholar |
|---|---|
| Author | Firmansyah, Achmad Fauzi Bagus Pramana, Setia |
| Copyright Year | 2018 |
| Abstract | Fuzzy Cluster is a clustering method that allows data to be a member of two or more clusters by combining hard-clustering method and fuzzy membership matrix. Two popular fuzzy clustering algorithms are Fuzzy C-Means (FCM) and Gustafson Kessel (GK). Although GK has better performance, GK has weakness handling linearly correlated data. Beside that, both FCM and GK produce unstable result due to randomization on parameters initialization. That weakness can be overcome by using improved covariance estimation and cluster ensemble, respectively. This study is aimed to implement cluster ensemble on fuzzy clustering (GK and FCM). The clustering performance between GK-Ensemble and FCM-Ensemble in generated dataset is investigated by using the Xie Beni index and missclassification rate. The results show that the GK-Ensemble outperform the FCM-Ensemble. The GK-Ensemble performs best in both case of overlapping clusters and well-separated clusters. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://commdis.telkomuniversity.ac.id/jdsa/index.php/jdsa/article/download/6/3 |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
