Multi-stage Hierarchical Clustering Method Based on Hypergraph

1Citations
Citations of this article
2Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Clustering analysis is a data analysis technique, it groups a set of data points into multiple clusters with similar data points. However, clustering of high dimensional data is still a difficult task. In order to facilitate this task, people usually use hypergraphs to represent the complex relationships between high dimensional data. In this paper, the hypergraph is used to improve the representation of the complex high dimensional data, and a multi-stage hierarchical clustering method based on hypergraph partition and Chameleon algorithm is proposed. The proposed method constructs a hypergraph in the shared-nearest-neighbor (SNN) graph from the dataset and then employs a hypergraph partitioning method hMETIS to obtain a series of subgraphs, finally those subgraphs are merged to get the final clusters. Experiments show that the proposed method is better than Chameleon algorithm and the other four clustering methods when applied on four UCI datasets.

Cite

CITATION STYLE

APA

Xi, Y., & Lu, Y. (2020). Multi-stage Hierarchical Clustering Method Based on Hypergraph. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 12465 LNAI, pp. 432–443). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-60796-8_37

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free