source: prdatasets/breast.m @ 151

Last change on this file since 151 was 150, checked in by bduin, 5 years ago
File size: 1.5 KB
RevLine 
[80]1%BREAST 699 objects with 9 features in 2 classes
2%
[94]3%       X = BREAST
[80]4%
5% Breast cancer Wisconsin dataset obtained from the University of Wisconsin
6% Hospitals, Madison from Dr. William H. Wolberg.
7%
8% REFERENCE
9% O. L. Mangasarian and W. H. Wolberg: "Cancer diagnosis via linear
10% programming", SIAM News, Volume 23, Number 5, September 1990, pp 1 & 18.
11%
[94]12%       X = BREAST(VAL)
[80]13%
[94]14% By default objects with missing values are removed. When something else
[137]15% is desired, use one of the options in MISVAL for VAL.
[94]16%
[150]17% SEE ALSO <a href="http://prtools.tudelft.nl/prtools/">PRTools Guide</a>, <a href="http://archive.ics.uci.edu/ml/">UCI Website</a>
[94]18% PRTOOLS, DATASETS, MISVAL
19
[150]20% Copyright: R.P.W. Duin
[94]21
[137]22function a = breast(val)
[80]23
[123]24if nargin < 1, val = 'remove'; end
[80]25
[137]26a = pr_loadmatfile;
27if isempty(a)
28  opt.delimeter = ',';
29  opt.labfeat   = 11;
30  opt.featnames = {'Clump Thickness' 'Uniformity of Cell Size' ...
31    'Uniformity of Cell Shape' 'Marginal Adhesion' ...
32    'Single Epithelial Cell Size' 'Bare Nuclei' 'Bland Chromatin' ...
33    'Normal Nucleoli' 'Mitoses'};
34  opt.feats = [2:10];
35  opt.misvalue = -1;
36  opt.classnames = {'benign' 'malignant'};
37  opt.desc='The original database of the Wisconsin Breast Cancer Databases from UCI, containing 699 instances, collected between 1989 and 1991. ';
38  opt.link = 'ftp://ftp.ics.uci.edu/pub/machine-learning-databases/breast-cancer-wisconsin/';
39  opt.dsetname = 'Breast Wisconsin';
40  a = pr_download('http://prtools.tudelft.nl/prdatasets/breastorg.dat',[],opt);
41end
42a = misval(a,val);
[123]43
[80]44return
Note: See TracBrowser for help on using the repository browser.