This is a modified version of a breast cancer data set [1] from the
UCI Machine Learing repository. The attributes are as follows:

   #  Attribute                     Domain
   -- -----------------------------------------
   1. Clump Thickness               low / high
   2. Uniformity of Cell Size       low / high
   3. Uniformity of Cell Shape      low / high
   4. Marginal Adhesion             low / high
   5. Single Epithelial Cell Size   low / high
   6. Bare Nuclei                   low / high
   7. Bland Chromatin               low / high
   8. Normal Nucleoli               low / high
   9. Mitoses                       low / high
  10. Class                         (2 for benign, 4 for malignant)

To load the data just run the script breast (breast.m).

[1] O. L. Mangasarian and W. H. Wolberg: "Cancer diagnosis via linear 
programming", SIAM News, Volume 23, Number 5, September 1990, pp 1 & 18.
