No Access

WEIGHTED NEIGHBORHOOD CLASSIFIER FOR THE CLASSIFICATION OF IMBALANCED TUMOR DATASET

SHU-LIN WANG

School of Computer and Communication, Hunan University, Changsha, Hunan 410082, China

Intelligent Computing Lab, Institute of Intelligent Machines, Chinese Academy of Science, Hefei, China

Search for more papers by this author

XUELING LI

Intelligent Computing Lab, Institute of Intelligent Machines, Chinese Academy of Science, Hefei, China

Search for more papers by this author

JUN-FENG XIA

Intelligent Computing Lab, Institute of Intelligent Machines, Chinese Academy of Science, Hefei, China

School of Life Science, University of Science and Technology of China, Hefei, Anhui 230027, China

Search for more papers by this author

, and

Xiao-Ping Zhang

Dept. of Electrical & Computer Engineering, Ryerson University, Toronto, Ontario, Canada, M5B 2K3 Canada

Corresponding author.

Search for more papers by this author

https://doi.org/10.1142/S0218126610006232Cited by:9 (Source: Crossref)

Abstract

Machine learning is widely applied to gene expression profiles based molecular tumor classification, but sample imbalance problem is often overlooked. This paper proposed a subclass-weighted neighborhood classifier to address the imbalanced sample set problem and a novel neighborhood rough set model to select informative genes for classification performance improvement. Experiments on three publicly available tumor datasets demonstrated that the proposed method is obviously effective on imbalanced dataset with obscure boundary between two subtypes and informative gene selection and it can achieve higher cross-validation accuracy with much fewer tumor-related genes.

Keywords:

References

M. Schena , DNA Microarrays: A Practical Approach ( Oxford University Press , New York , 2000 ) . Google Scholar
L. M. Fu and C. S. Fu-Liu, FEBS Lett. 561, 186 (2004), DOI: 10.1016/S0014-5793(04)00175-9. Crossref, Web of Science, Google Scholar
C. H. Zhanget al., IEEE Tran. Inform. Tech. Biomed. 13, 599 (2009). Web of Science, Google Scholar
D. G. Covellet al., Mol. Cancer Therap. 2, 317 (2003). Web of Science, Google Scholar
J. Khanet al., Nat. Med. 7, 673 (2001), DOI: 10.1038/89044. Crossref, Web of Science, Google Scholar
I. Guyonet al., Mach. Learn. 46, 389 (2002), DOI: 10.1023/A:1012487302797. Crossref, Web of Science, Google Scholar
J. W. Fang and J. W. Grzymala-Busse, LNCS 4029 (2006) pp. 899–908. Google Scholar
L. Liet al., Bioinformatics 17, 1131 (2001), DOI: 10.1093/bioinformatics/17.12.1131. Crossref, Web of Science, Google Scholar
K. Yanget al., Evaluation models for the effect of sample imbalance on gene selection, Proc. 1st Int. Multi-Symp. Computer and Computational Sciences (2006) pp. 56–63. Google Scholar
D. S. Huang and C. H. Zheng, Bioinformatics 22, 1855 (2006), DOI: 10.1093/bioinformatics/btl190. Crossref, Web of Science, Google Scholar
H. L. Huang, C. C. Lee and S. Y. Ho, BioSystems 90, 78 (2007), DOI: 10.1016/j.biosystems.2006.07.002. Crossref, Web of Science, Google Scholar
M. M. Xiong, X. Z. Fang and J. Y. Zhao, Genome Res. 11, 1878 (2001). Crossref, Web of Science, Google Scholar
Z. X. Zhu, Y. S. Ong and M. Dash, Pattern Recogn. 40, 3236 (2007), DOI: 10.1016/j.patcog.2007.02.007. Crossref, Web of Science, Google Scholar
Z. Pawlak, Int. J. Comput. Info. Sci. 11, 341 (1982), DOI: 10.1007/BF01001956. Crossref, Google Scholar
K. Thangavel and A. Pethalakshmi, Appl. Soft Comput. 9, 1 (2009), DOI: 10.1016/j.asoc.2008.05.006. Crossref, Web of Science, Google Scholar
H. Midelfartet al., Fundamenta Informaticae 53, 155 (2002). Web of Science, Google Scholar
J. W. Fang and J. W. Grzymala-Busse, LNCS 4062 (2006) pp. 758–765. Google Scholar
Q. Hu, D. Yu and Z. Xie, J. Software 19, 640 (2008), DOI: 10.3724/SP.J.1001.2008.00640. Crossref, Web of Science, Google Scholar
Q. Hu, D. Yu and Z. Xie, Expert Syst. Appl. 34, 866 (2008), DOI: 10.1016/j.eswa.2006.10.043. Crossref, Web of Science, Google Scholar
S. Tan, Expert Syst. Appl. 28, 667 (2005), DOI: 10.1016/j.eswa.2004.12.023. Crossref, Web of Science, Google Scholar
W. H. Kruskal and W. A. Wallis, J. Amer. Stat. Assoc. 47, 583 (1952), DOI: 10.2307/2280779. Crossref, Web of Science, Google Scholar
T. R. Golubet al., Science 286, 531 (1999), DOI: 10.1126/science.286.5439.531. Crossref, Web of Science, Google Scholar
U. Alonet al., PNAS. USA 96, 6745 (1999), DOI: 10.1073/pnas.96.12.6745. Crossref, Web of Science, Google Scholar
J. Deutsch, Bioinformatics 19, 45 (2003), DOI: 10.1093/bioinformatics/19.1.45. Crossref, Web of Science, Google Scholar
I. Wadmanet al., EMBO J. 13, 4831 (1994). Crossref, Web of Science, Google Scholar
S. J. Orret al., Blood 109, 1061 (2007), DOI: 10.1182/blood-2006-05-023556. Crossref, Web of Science, Google Scholar
N. Niitsuet al., Leukemia 17, 196 (2003), DOI: 10.1038/sj.leu.2402699. Crossref, Web of Science, Google Scholar
M. Tatsutaet al., Cancer Lett. 93, 219 (1995), DOI: 10.1016/0304-3835(95)03813-C. Crossref, Web of Science, Google Scholar
O. Hillet al., Bioch. Bioph. Acta 1253, 146 (1995). Web of Science, Google Scholar
K. Klimek-Tomczaket al., Brit. J. Cancer 94, 586 (2006), DOI: 10.1038/sj.bjc.6602938. Crossref, Web of Science, Google Scholar
T. S. Fureyet al., Bioinformatics 16, 906 (2000), DOI: 10.1093/bioinformatics/16.10.906. Crossref, Web of Science, Google Scholar
B. Krishnapuram, L. Carin and A. Hartemink, Kernel Methods in Computational Biology, eds. B. Schölkopf, K. Tsuda and J.-P. Vert (MIT Press, Cambridge, MA, 2004) pp. 299–317. Crossref, Google Scholar
D. V. Nguyen and D. M. Rocke, Bioinformatics 18, 39 (2002), DOI: 10.1093/bioinformatics/18.1.39. Crossref, Web of Science, Google Scholar
S. L. Wanget al., Gene selection with rough sets for the molecular diagnosing of tumor based on support vector machines, Int. Computer Symp. 2006 (2006) pp. 1368–1373. Google Scholar
H. Q. Wang and D. S. Huang, Pattern Recogn. Lett. 27, 116 (2006), DOI: 10.1016/j.patrec.2005.07.007. Crossref, Web of Science, Google Scholar
B. Haibe-Kainset al., Bioinformatics 24, 2200 (2008), DOI: 10.1093/bioinformatics/btn374. Crossref, Web of Science, Google Scholar