BulDML at Institute of Mathematics and Informatics >
IMI Periodicals >
Serdica Journal of Computing >
2014 >
Volume 8 Number 2 >

Please use this identifier to cite or link to this item:

Title: A Comparative Analysis of Predictive Learning Algorithms on High-Dimensional Microarray Cancer Data
Authors: Bill, Jo
Fokoue, Ernest
Keywords: HDLSS
Machine Learning Algorithm
Pattern Recognition
Discriminant Analysis
Support Vector Machine
Cross Validation
Microarray Cancer Data
Issue Date: 2014
Publisher: Institute of Mathematics and Informatics Bulgarian Academy of Sciences
Citation: Serdica Journal of Computing, Vol. 8, No 2, (2014), 137p-168p
Abstract: This research evaluates pattern recognition techniques on a subclass of big data where the dimensionality of the input space (p) is much larger than the number of observations (n). Specifically, we evaluate massive gene expression microarray cancer data where the ratio κ is less than one. We explore the statistical and computational challenges inherent in these high dimensional low sample size (HDLSS) problems and present statistical machine learning methods used to tackle and circumvent these difficulties. Regularization and kernel algorithms were explored in this research using seven datasets where κ < 1. These techniques require special attention to tuning necessitating several extensions of cross-validation to be investigated to support better predictive performance. While no single algorithm was universally the best predictor, the regularization technique produced lower test errors in five of the seven datasets studied.
ISSN: 1312-6555
Appears in Collections:Volume 8 Number 2

Files in This Item:

File Description SizeFormat
sjc-vol8-num2-2014-p137-p168.pdf428.3 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0!   Creative Commons License