A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs
SIAM Journal on Scientific Computing
Cluster ensembles --- a knowledge reuse framework for combining multiple partitions
The Journal of Machine Learning Research
The Journal of Machine Learning Research
Solving cluster ensemble problems by bipartite graph partitioning
ICML '04 Proceedings of the twenty-first international conference on Machine learning
Clustering Ensembles: Models of Consensus and Weak Partitions
IEEE Transactions on Pattern Analysis and Machine Intelligence
ACM Transactions on Knowledge Discovery from Data (TKDD)
Moderate diversity for better cluster ensembles
Information Fusion
Solving Consensus and Semi-supervised Clustering Problems Using Nonnegative Matrix Factorization
ICDM '07 Proceedings of the 2007 Seventh IEEE International Conference on Data Mining
Generalized cluster aggregation
IJCAI'09 Proceedings of the 21st international jont conference on Artifical intelligence
Graph Regularized Nonnegative Matrix Factorization for Data Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
A self-supervised framework for clustering ensemble
WAIM'13 Proceedings of the 14th international conference on Web-Age Information Management
Hi-index | 0.00 |
Cluster ensembles aim to generate a stable and robust consensus clustering by combining multiple different clustering results of a dataset. Multiple clusterings can be represented either by multiple co-association pairwise relations or cluster based features. Traditional clustering ensemble algorithms learn the consensus clustering using either of the two representations, but not both. In this paper, we propose to integrate the two representations in a unified framework by means of weighted graph regularized nonnegative matrix factorization. Such integration makes the two representations complementary to each other and thus outperforms both of them in clustering accuracy and stability. Extensive experimental results on a number of datasets further demonstrate this.