| Parity of bits 2,3,4,6,8 (1,5,7,9,10 irrelevant).  By Ronny Kohavi
| First used in John, Kohavi, and Pfleger, 
|    Irrelevant features and the subset selection
|    problem. In Machine Learning: Proceedings of the Eleventh International
|    Conference, 1994, available in http://robotics.stanford.edu/~ronnyk
| The training set gives 100 randomly selected instances.
|   Induction algorithms doing good feature subset selection thus
|   see all instances (there are only 32 instances over 5 bits).

0, 1.

Bit 1: 0, 1.
Bit 2: 0, 1.
Bit 3: 0, 1.
Bit 4: 0, 1.
Bit 5: 0, 1.
Bit 6: 0, 1.
Bit 7: 0, 1.
Bit 8: 0, 1.
Bit 9: 0, 1.
Bit 10: 0, 1.