A Comparative Study of Efficient Initialization Methods for the K-Means Clustering Algorithm

Explore this paper's citation graph

Summary

It is demonstrated that popular initialization methods often perform poorly and that there are in fact strong alternatives to these methods, and eight commonly used linear time complexity initialization methods are compared.

Type
article
Published
2012-09-10
Cited by
1,226
References
89
Access
Open access

Keywords

Initialization, Computer science, Cluster analysis, Gradient descent, Algorithm

References

Cited by

Related papers