SOM2CE: double self-organizing map based cluster ensemble framework and its application in cancer gene expression profiles

Authors:
Zhiwen Yu;Hantao Chen;Jane You;Le Li;Guoqiang Han
Affiliations:
School of Computer Science and Engineering, South China University of Technology, China,Department of Computing, Hong Kong Polytechnic University, Hong Kong;School of Computer Science and Engineering, South China University of Technology, China;Department of Computing, Hong Kong Polytechnic University, Hong Kong;School of Computer Science and Engineering, South China University of Technology, China;School of Computer Science and Engineering, South China University of Technology, China
Venue:
IEA/AIE'12 Proceedings of the 25th international conference on Industrial Engineering and Other Applications of Applied Intelligent Systems: advanced research in applied artificial intelligence
Year:
2012

Citing 6
Cited 0

Normalized Cuts and Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Voting-Merging: An Ensemble Method for Clustering

ICANN '01 Proceedings of the International Conference on Artificial Neural Networks
Adaptive Clustering Ensembles

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 1 - Volume 01
Clustering Ensembles: Models of Consensus and Weak Partitions

IEEE Transactions on Pattern Analysis and Machine Intelligence
Multi-objective clustering ensemble

International Journal of Hybrid Intelligent Systems - Hybridization of Intelligent Systems
Hybrid cluster ensemble framework based on the random combination of data transformation operators

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

Though there exist a lot of cluster ensemble approaches, few of them consider how to degrade the effect of noisy attributes in the dataset. In the paper, we propose a new cluster ensemble framework, named as double self-organizing map based cluster ensemble (SOM2CE) to perform clustering on noisy datasets. SOM2CE incorporates the self-organizing map (SOM) twice into the ensemble framework to discovery the underlying structure of noisy datasets, which applies SOM to perform clustering not only on the sample dimension, but also on the attribute dimension. SOM2CE also adopts the normalized cut algorithm to partition the consensus matrix constructed from multiple clustering solutions, and obtain the final results. Experiments on both synthetic datasets and cancer gene expression profiles illustrate that the proposed approach not only achieves good performance on synthetic datasets and cancer gene expression profiles, but also outperforms most of the existing approaches in the process of clustering gene expression profiles.