The optimal k-means clustering on Z is the same as that on V , if any intra-cluster transitive distance is less than half of any inter-cluster transitive distance. We can prove under such situations the following in- equality holds: d(zi,zk) > d(zi,zj) where i, j belong to the same cluster and k, i different clusters.
Nov 22, 2007 · In this paper, a non-eigenproblem based clustering method is proposed to deal with the clustering problem. Its performance is comparable to the ...
Specifically, by involving a more realistic and effective distance and the "k-means duality" property, our algorithm can handle datasets with complex cluster ...
We show that with a transitive distance and an observed property, called K-means duality, our algorithm can be used to handle data sets with complex cluster ...
It is shown that with a transitive distance and an observed property, called K-means duality, the proposed non-eigenproblem based clustering method can be ...
Given a weighted graph with edge weights, each transitive edge lies on the minimum spanning tree. The K-Means Duality. Denote: V the set of data. E the ...
Specifically, by involving a more realistic and effective distance and the "k-means duality" property, our algorithm can handle datasets with complex cluster ...
Nov 22, 2007 · Algorithm 1 Clustering Based on the Transitive Distance and the K-means Duality. 1) Construct a weighted complete graph G = (V,E) where E ...
while capable of gaining better performance in our ex- periments. Specifically, by involving a more realistic and effective distance and the “k-means ...
Specifically, by involving a more realistic and effective distance and the "k-means duality" property, our algorithm can handle datasets with complex cluster ...