|
Class Summary |
| BagCorpusCounter |
A corpus counter implemented as a collection of bags. |
| ChiSquared |
A SimpleOrderer that orders according to Pearson's Chi-Squared test. |
| ClassificationOrderer |
An Orderer that orders according to usefullness during classification. |
| CorpusCounter |
Assume a sampling process, where each sample determines two discrete random
variables, C and W. |
| Method0 |
A TwoLevelOrderer that combines the first level score with the second
level scores according to the following formula:
firstLevelScore / Sum_c_in_categories (seconedLevelScore_c)
This orderer ignores its parameter. |
| Method3 |
A TwoLevelOrderer that combines the first level score with the second
level scores according to the following formula:
<>
This orderer uses its double parameter in its calculation. |
| MutualInformation |
A SimpleOrderer that orders according to the following formula:
score (w) = sum_c_in_C_and_w_in_W [p(c,w) * log [p(c,w) / p(c) / p(w)]]
where C is the set of categories and V is {^w,w}. |
| NegativeMutualInformation |
A SimpleOrderer that orders according to the following formula:
score (w) =
-1 * sum_c_in_C_and_w_in_W [p(c,w) * log [p(c,w) / p(c) / p(w)]]
where C is the set of categories and V is {w,^w}. |
| None |
An Orderer that does not change the order of the potential features. |
| PointwiseChiSquared |
A SimpleOrderer that orders according to according to the maximum of the
constiuent addends in Pearson's Chi-Squared test. |
| PointwiseMutualInformation |
A SimpleOrderer that orders according to the following formula:
score (w) = max_c_in_C_and_w_in_W [log [p(c,w)/p(c)/p(w)]]
where C is the set of categories and W is {^w, w}. |
| Random |
A SimpleOrderer that orders randomly. |
| Reducer |
Provides a routine that eliminates from a set of potential features those
potential features that occur too seldom or too often. |
| Selector |
Provides a routine that, given a ordered set of potential features, selects
the best. |
| SimpleOrderer |
A ClassificationOrderer that orders from the corpus counter alone. |
| TwoLevelOrderer |
A ClassificationOrderer that orders from the corpus counter and the
category counters. |
| TwoLevelScore.ReverseScore1 |
The following Comparator orders TwoLevelScores in reverse order
of their first level scores (not their combined scores as the comparator
in DoubleWrap would). |
| Util |
Provides some utility routines. |