Title :
FTCluster: Efficient Mining Fault-Tolerant Biclusters in Microarray Dataset
Author :
Wang, Miao ; Shang, Xuequn ; Miao, Miao ; Li, Zhanhuai ; Liu, Wenbin
Author_Institution :
Sch. of Comput., Northwestern Polytech. Univ., Xi´´an, China
Abstract :
Biclustering is a popular method for micro array dataset analysis. It allows for condition set and gene set points clustering simultaneously. However, the noisy data in micro array may disturb the mining results. In order to reduce the influence of noise and find more biological biclusters, we propose an algorithm, FT Cluster, to mine fault-tolerant biclusters in micro array dataset. Unlike traditional fault-tolerant biclusters mining algorithms, FT Cluster uses several novel techniques to improve the efficiency. It also adopts several techniques to generate relaxed biclusters without candidate maintenance. The experimental results show FT Cluster is more effective than traditional algorithms. The biological significance of FT Cluster is evaluated by Gene Ontology and the results show FT Cluster can find larger biological relevant biclusters.
Keywords :
biology computing; data mining; fault tolerant computing; genetics; lab-on-a-chip; ontologies (artificial intelligence); pattern clustering; FTCluster; biological biclusters; condition set point clustering; fault tolerant bicluster mining; gene ontology; gene set point clustering; microarray dataset analysis; noisy data; Algorithm design and analysis; Data mining; Fault tolerance; Fault tolerant systems; Gene expression; Maintenance engineering; biclustering; fault-tolerant bicluster; gene expression; microarray;
Conference_Titel :
Data Mining Workshops (ICDMW), 2011 IEEE 11th International Conference on
Conference_Location :
Vancouver, BC
Print_ISBN :
978-1-4673-0005-6
DOI :
10.1109/ICDMW.2011.89
