یک الگوریتم خوشه بندی مبتنی بر مورچه جدید با استفاده از روش هسته
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|79088||2011||15 صفحه PDF||سفارش دهید||8323 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Information Sciences, Volume 181, Issue 20, 15 October 2011, Pages 4658–4672
A novel ant-based clustering algorithm integrated with the kernel (ACK) method is proposed. There are two aspects to the integration. First, kernel principal component analysis (KPCA) is applied to modify the random projection of objects when the algorithm is run initially. This projection can create rough clusters and improve the algorithm’s efficiency. Second, ant-based clustering is performed in the feature space rather than in the input space. The distance between the objects in the feature space, which is calculated by the kernel function of the object vectors in the input space, is applied as a similarity measure. The algorithm uses an ant movement model in which each object is viewed as an ant. The ant determines its movement according to the fitness of its local neighbourhood. The proposed algorithm incorporates the merits of kernel-based clustering into ant-based clustering. Comparisons with other classic algorithms using several synthetic and real datasets demonstrate that ACK method exhibits high performance in terms of efficiency and clustering quality.