hypex - computes the Chernoff exponent between two simple categories.
hypex [-hH size] [-s stepsize] CATDUMP1 CATDUMP2
hypex reads two category dumps produced by dbacl(1) after learning. A
category dump is obtained using the -d and -l switches, and is a
textual representation of the feature weights which exist in the binary
Given two such category dumps for simple unigram categories, hypex
calculates the Kullback Leibler divergence between the probability
models, and prints out exponential error exponents for Neyman-Pearson
hypothesis tests under a range of threshold values. See Cover and
Thomas (1991) Elements of Information Theory, Chap. 12.
Note that only simple categories are supported, and the output only
makes sense under appropriate theoretical conditions. This is a very
hypex returns 0 on success, 1 if an error occurs.
-H Same as dbacl(1). Selects the hash sizes in powers of two.
-s Stepsize for the threshold. hypex outputs exponents for
different values of the threshold, within an interval bounded by
the Kullback Leibler divergences between the categories.
-V Print the program version number and exit.
The source code for the latest version of this program is available at
the following locations:
Doesn’t work with complex categories, and theoretical assumptions are
unrealistic in practice.
Laird A. Breyer <email@example.com>
dbacl(1), mailcross(1), mailfoot(1), mailinspect(1), mailtoe(1),