k-Means is a very popular clustering algorithm. We modify its objective to achieve a clustering method which produces more balanced clusters. The proposal can be adapted in a framework where dataset keeps growing and number of clusters is decided within the algorithm to achieve balanced clustering. This is done without affecting the time complexity. Experimental results are in favor of the proposal.
CITATION STYLE
Saini, D., Singh, M., & Sharma, I. (2016). Variance-based clustering for balanced clusters in growing datasets. In Advances in Intelligent Systems and Computing (Vol. 438, pp. 559–565). Springer Verlag. https://doi.org/10.1007/978-981-10-0767-5_58
Mendeley helps you to discover research relevant for your work.