Tumor clustering based on hybrid cluster ensemble framework

Zhiwen Yu, Jia You, Hantao Chen, Le Li, Xiaowei Wang

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

1 Citation (Scopus)


Tumor clustering from bio-molecular data provides a new way to perform cancer class discovery. In this paper, we propose a hybrid fuzzy cluster ensemble framework (HFCEF) for tumor clustering from cancer gene expression data. Compared with traditional cluster ensemble framework, HFCEF integrates both the hard clustering and the soft clustering into the cluster ensemble framework. Specifically, HFCEF first applies the affinity propagation algorithm (AP) to perform clustering on the attribute dimension, and generates a set of subspaces which are used to create a set of new datasets. Then, the fuzzy membership function and the affinity propagation algorithm are adopted to generate a set of fuzzy matrices in the ensemble. Finally, the normalized cut algorithm is served as the consensus function to summarize the set of fuzzy matrices and obtain the final result. The experiments on cancer gene expression profiles shows that the proposed framework works well on bio-molecular data, and provides more robust, stable and accurate results.
Original languageEnglish
Title of host publicationICCH 2012 Proceedings - International Conference on Computerized Healthcare
PublisherIEEE Computer Society
Number of pages7
ISBN (Print)9781467351294
Publication statusPublished - 1 Jan 2012
Event2012 International Conference on Computerized Healthcare, ICCH 2012 - Hong Kong, Hong Kong
Duration: 17 Dec 201218 Dec 2012


Conference2012 International Conference on Computerized Healthcare, ICCH 2012
Country/TerritoryHong Kong
CityHong Kong

ASJC Scopus subject areas

  • Health Informatics


Dive into the research topics of 'Tumor clustering based on hybrid cluster ensemble framework'. Together they form a unique fingerprint.

Cite this