The ols (Orthogonal Least Squares) program transforms each example from a data file into a fuzzy rule, and selects the most important rules according to a least square criterion. It uses linear regression and Gram-Schmidt orhtogonalization. Once the rules selected, a second passage of the algorithm is done to optimize the rule conclusions. This algorithm is well suited to regression problems.
Java interface:
Learning menu, Rule induction submenu, OLS option.
Command line, ols program:
A single argument:
Options:
i=0: the new rule conclusions will be chosen according to the data file
i=1 (default value): the new rule conclusions will be chosen according to the initial rule conclusions.
Command line example:
ols rice
The program creates 5 files:
Num Index VarExp VarCum 1, 18, 0.392840, 0.392840, 2, 102, 0.197165, 0.590004, 3, 85, 0.116578, 0.706582, 4, 8, 0.088941, 0.795523, 5, 64, 0.031271, 0.826794, 6, 31, 0.023540, 0.850334, 7, 25, 0.015977, 0.866311, 8, 100, 0.014835, 0.881146, 9, 20, 0.014392, 0.895538, 10, 49, 0.013351, 0.908889, 11, 28, 0.012675, 0.921563, 12, 35, 0.009627, 0.931191, 13, 1, 0.009520, 0.940711, 14, 2, 0.014960, 0.955671, 15, 42, 0.006941, 0.962613, 16, 16, 0.006291, 0.968903, 17, 34, 0.004313, 0.973216, 18, 71, 0.004237, 0.977453, 19, 13, 0.002217, 0.979671, 20, 79, 0.001903, 0.981573, 21, 53, 0.001852, 0.983426, 22, 14, 0.001719, 0.985144, 23, 4, 0.001735, 0.986879, 24, 12, 0.002596, 0.989475, 25, 10, 0.001168, 0.990643,A line per induced rule.
For each line, the following columns: