A New Descriptor Selection Scheme for SVM in Unbalanced Class Problem: A Case Study Using Skin Sensitisation Dataset

Details

• Personal Author:
  Andrew, Michael E. ; Fedorowicz A ; Li S
• Description:
  A novel descriptor selection scheme for Support Vector Machine (SVM) classification method has been proposed and its utility demonstrated using a skin sensitisation dataset as an example. A backward elimination procedure, guided by mean accuracy (the average of specificity and sensitivity) of a leave-one-out cross validation, is devised for the SVM. Subsets of descriptors were first selected using a sequential t-test filter or a Random Forest filter, before backward elimination was applied. Different kernels for SVM were compared using this descriptor selection scheme. The Radial Basis Function (RBF) kernel worked best when a sequential t-test filter was adopted. The highest mean accuracy, 84.9%, was obtained using SVM with 23 descriptors. The sensitivity and the specificity were as high as 93.1% and 76.6%, respectively. A linear kernel was found to be optimal when a Random Forest filter was used. The performance using 24 descriptors was comparable with a RBF kernel with a sequential t-test filter. As a comparison, Fisher's linear discriminant analysis (LDA) under the same descriptor selection scheme was carried out. SVM was shown to outperform the LDA. [Description provided by NIOSH]
• Subjects:
  Occupational Health Safety Skin
• Keywords:
  Analytical-methods Analytical-processes Mathematical-models Skin-irritants Skin-sensitivity Statistical-analysis
• ISSN:
  1062-936X
• Document Type:
  Text
• Genre:
  Journal Article
• Place as Subject:
  OSHA Region 3 ; West Virginia
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Division:
  HELD - Health Effects Laboratory Division
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Pages in Document:
  423-441
• Volume:
  18
• Issue:
  5
• NIOSHTIC Number:
  nn:20032635
• Citation:
  SAR QSAR Environ Res 2007 Jul; 18(5-6):423-441
• Contact Point Address:
  S. Li, Health Effects Laboratory Division, National Institute for Occupational Safety and Health, Morgantown, WV 26505
• Email:
  swl4@cdc.gov
• Federal Fiscal Year:
  2007
• Peer Reviewed:
  True
• Source Full Name:
  SAR and QSAR in Environmental Research
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:92e8372aec277adbf965d0f7237dcd5907ca025b23a76ade165fac2d03857ae36946faf8dcca8e2ffc12178a6fd76c3a98a72ed12f41c4cda8465d4534d122f7
• Download URL:
  https://stacks.cdc.gov/view/cdc/186535/cdc_186535_DS1.pdf
• File Type:
  [PDF - 1.64 MB ]