Careful Seeding Based on Independent Component Analysis for k-Means Clustering
Takashi Onoda, Miho Sakai, Seiji Yamada · 2010
The k-means method is a widely used clustering technique because of its simplicity and speed. However, the clustering result depends heavily on the chosen initial value. In this report, we propose a seeding method with independent component analysis for the k-means method. Using a benchmark dataset, we evaluate the performance of our proposed method and compare it with other seeding methods.