bayesol.1

dbacl是一个通用目的的digramic贝叶斯文本分类器。它可以学习你提供的文本

\" t
.TH BAYESOL 1 "Bayesian Classification Tools" "Version 1.3" ""
.SH NAME
bayesol \- a Bayes solution calculator for use with dbacl.
.SH SYNOPSIS
.HP
.B bayesol
[-DVniv] -c 
.I riskspec
[FILE]...
.HP
.B bayesol
-V
.SH DESCRIPTION
.PP
.B bayesol
is a Bayes solution calculator designed to combine the output of 
.BR dbacl (1) 
with a prior distribution and a risk specification, and
calculate the optimal Bayesian decision (which minimizes the posterior
risk). 
.PP
The risk specification is read from the text file 
.I riskspec 
and must be written in a simple format described below. The 
.BR dbacl (1) 
output can either be read from FILE or from STDIN. 
.SH EXIT STATUS
On success, 
.B bayesol
returns a positive integer corresponding to the category with the lowest risk.
In case of a problem, 
.B bayesol
returns zero.
.SH OPTIONS
.IP -c
Classify using 
.IR riskspec . 
See the section RISK SPECIFICATION.
.IP -i
Fully internationalized mode. Forces the use of wide characters internally,
which is necessary in some locales. This incurs a noticeable performance penalty.
.IP -n
Print risk scores for each 
.IR category . 
Each score is the logarithm of the expected risk under that category.
.IP -v
Verbose mode. Prints to STDOUT the category with minimum posterior risk.
In case several categories are possible, 
prints the first category in the order in which they appear
in the categories section of 
.IR riskpspec .
.IP -D
Print debug output. Do not use.
.IP -V
Print the program version number and exit. 
.SH RISK SPECIFICATION
.B bayesol
needs to read a text file 
.I riskspec
containing a risk specification. The format of this text file is as follows
.IP
.na
categories { 
.IR cat1 , 
.IR cat2 , "" ..., 
.IR catN }
.br
prior { 
.IR p1 , 
.IR p2 , "" ..., 
.IR pN }
.br
loss_matrix {
.br
"\fIregex1\fR" \fIc1\fR [ 
.IR formula11 ,
.IR formula12 , "" ...,
.IR formula1N ]
.br
"\fIregex2\fR" \fIc2\fR [ 
.IR formula21 ,
.IR formula22 , "" ...,
.IR formula2N ]
.br 
 .
.br
 .
.br
"\fIregexM\fR" \fIcM\fR [ 
.IR formulaM1 , 
.IR formulaM2 , "" ..., 
.IR formulaMN ]
.br
}
.br
.ad
.PP
In the above, 
.IR cat1 ,
.IR cat2 , "" ..., 
.IR catN , 
are category names, 
.IR p1 ,
.IR p2 , "" ...,
.IR pN ,
are non-negative numbers, 
.IR regex1 ,
.IR regex2 , "" ...,
.IR regexM ,
are (possibly empty) regular expression strings, 
.IR c1 ,  
.IR c2 , "" ...,
.IR cM ,
are instances of the category names 
.IR cat1 ,
.IR cat2 , "" ...,
.IR catN , 
and the formulas are numbers or mathematical expressions. 
.PP
Every category which appears in the categories section must appear at least
once in the loss_matrix section, with an empty "" regular expression.
To construct the actual loss matrix used in the decision calculations, 
.B bayesol 
selects, for each category appearing in the categories section,
the first row whose regular expression is matched
within FILE or STDIN, or the first row with empty regular expression if there
are no matches.
.PP
Each formula can be either a single number, or an algebraic combination of
the operators exp(), log(), +, -, *, /, ^ and parentheses (). The string "inf"
is parsed as the value infinity. Also, the 
string "complexity" is recognized, and converted to the complexity for 
that category 
as reported by 
.BR dbacl (1).
Finally, if the 
corresponding regular expression contains submatches delimited by parentheses, 
their numerical values can be used inside the formulas as the special variables
$1, ..., $9. Note that submatches which aren't numerical are converted to the value zero.
.PP
Case is important. Spaces and newlines can be liberally inserted. Comments 
must start with a # and extend to the end of the line. 
.SH USAGE
.PP
Typically, 
.B bayesol 
is used together with 
.BR dbacl (1). 
An invocation looks like this:
.PP
.na
% dbacl -c one -c two -c three sample.txt -vna | bayesol -c toy.risk -v
.ad
.PP
See also
/usr/share/dbacl/doc/tutorial.html for a more detailed overview.
.SH SOURCE
.PP
The source code for the latest version of this program is available at 
http://www.lbreyer.com/gpl.html
.SH AUTHOR
.PP
Laird A. Breyer <[e-mail]>
.SH SEE ALSO
.PP
.BR dbacl (1), 
.BR mailcross (1),
.BR mailinspect (1),
.BR regex (7)