DocumentCode :
2075327
Title :
Towards Better Outliers Detection for Gene Expression Datasets
Author :
Kashef, R. ; Kamel, M.S.
Author_Institution :
Dept. of Electr. & Comput. Eng., Waterloo Univ., Waterloo, ON
fYear :
2008
fDate :
June 29 2008-July 5 2008
Firstpage :
149
Lastpage :
154
Abstract :
This paper compares the performance of three clustering algorithms on the task of outlier´s detection. The goal is to illustrate that better clustering indicates better detection of outliers. k-means (KM), Bisecting k-means (BKM) and the partitioning around medoids (PAM) algorithms are each combined with the clustering-based outliers detection (Find CBLOF) method. Undertaken experimental results over four gene expression datasets where outliers are presented show that the clustering solutions of the PAM algorithm enable the Find CBLOF algorithm to discover more outliers than those of both the k-means and the bisecting k-means algorithms. The main reason for this is that PAM provides better clustering quality than that of the other two clustering algorithms on the tested datasets measured by external and internal quality measures.
Keywords :
biology computing; genetics; pattern clustering; statistical analysis; bisecting k-means clustering; clustering algorithms; clustering-based outliers detection method; gene expression datasets; k-means clustering; partitioning around medoids algorithms; Bioinformatics; Biomedical computing; Biomedical engineering; Clustering algorithms; Clustering methods; Data engineering; Diseases; Gene expression; Partitioning algorithms; Pattern analysis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Biocomputation, Bioinformatics, and Biomedical Technologies, 2008. BIOTECHNO '08. International Conference on
Conference_Location :
Bucharest
Print_ISBN :
978-0-7695-3191-5
Electronic_ISBN :
978-0-7695-3191-5
Type :
conf
DOI :
10.1109/BIOTECHNO.2008.29
Filename :
4561150
Link To Document :
بازگشت