data(letter.recognition)
[,1] | lettr | capital letter |
[,2] | x.box | horizontal position of box |
[,3] | y.box | vertical position of box |
[,4] | width | width of box |
[,5] | high | height of box |
[,6] | onpix | total number of on pixels |
[,7] | x.bar | mean x of on pixels in box |
[,8] | y.bar | mean y of on pixels in box |
[,9] | x2bar | mean x variance |
[,10] | y2bar | mean y variance |
[,11] | xybar | mean x y correlation |
[,12] | x2ybr | mean of x^2 y |
[,13] | xy2br | mean of x y^2 |
[,14] | x.ege | mean edge count left to right |
[,15] | xegvy | correlation of x.ege with y |
[,16] | y.ege | mean edge count bottom to top |
[,17] | yegvx | correlation of y.ege with x |
The research for this article investigated the ability of several variations of Holland-style adaptive classifier systems to learn to correctly guess the letter categories associated with vectors of 16 integer attributes extracted from raster scan images of the letters. The best accuracy obtained was a little over 80%. It would be interesting to see how well other methods do with the same data.