Abstract | ||
---|---|---|
•A soft-balanced clustering framework is proposed in this paper, in which a novel regular objective is adopted to meet the balance requirement, and the balance of clustering can be flexibly adjusted by a parameter.•We skillfully apply this framework to the k-means algorithm to propose a balanced k-means with novel constraint (BKNC) model, which improves the balance of clustering results while retaining the efficient and convenient characteristics of k-means.•Extensive experimental results on real-world datasets show that the proposed BKNC outperforms other state-of-the-art hard-balanced and soft-balanced k-means methods. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1016/j.sigpro.2022.108572 | Signal Processing |
Keywords | DocType | Volume |
Clustering,Balanced clustering,Soft-balanced clustering,Balance regularizer | Journal | 198 |
ISSN | Citations | PageRank |
0165-1684 | 0 | 0.34 |
References | Authors | |
0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Huimin Chen | 1 | 0 | 0.34 |
Qianrong Zhang | 2 | 0 | 0.34 |
Rong Wang | 3 | 0 | 0.34 |
Feiping Nie | 4 | 0 | 0.34 |
Xuelong Li | 5 | 15049 | 617.31 |