.TH apertium-lextor 1 2006-12-12 "" ""
.SH NAME
apertium-lextor \- This application is part of
(
.B apertium
)
.PP
This tool is part of the apertium machine translation
architecture: \fBhttp://apertium.org\fR.
.SH SYNOPSIS
.B apertium-lextor
.B \-\-trainwrd\fR stopwords words n left right corpus model
[
.B \-\-weightexp\fR w
]
[
.B \-\-debug\fR
]
.PP
.B apertium\-lextor
.B \-\-trainlch\fR stopwords lexchoices n left right corpus wordmodel dic bildic model
[
.B \-\-weightexp\fR w
]
[
.B \-\-debug\fR
]
.PP
.B apertium\-lextor
.B \-\-lextor\fR model dic left right
[
.B \-\-debug\fR
]
[
.B \-\-weightexp\fR w
]
.PP
.SH DESCRIPTION
.BR apertium\-lextor 
is the application responsible for training and usage of the \fIlexical
selector module\fR.
.SH OPTIONS
.PP
.B \-\-trainwrd | \-t\fR
.br
Train word co-occurrences model. It needs the following required parameters:
.TP
.I stopwords\fR file containing a list of stop words. Stop words are ignored.
.TP
.I words\fR file containing a list of words. For each word a co-occurrence model is built.
.TP
.I n\fR number of words per co\-occurrence model (for each model, the \fIn\fR most frequent words).
.TP
.I left\fR left\-side context to take into account (number of words).
.TP
.I right\fR right\-side context to take into account (number of words).
.TP
.I corpus\fR file containing the training corpus.
.TP
.I model\fR output file on which the co\-occurrence models are saved.
.PP
.B \-\-trainlch | \-r\fR
.br
Train lexical choices co\-occurrence models using a target language
co\-occurrence model and a bilingual dictionary. It needs the
following required parameters:
.TP
.I stopwords\fR file containing a list of stop words. Stop words are ignored.
.TP
.I lexchoices\fR file containing a list of lexical choices. For each lexical choice a co\-occurrence model is built.
.TP
.I n\fR number of words per co\-occurrence model (for each model, the n most frequent words).
.TP
.I left\fR left\-side context to take into account (number of words).
.TP
.I right\fR right\-side context to take into account (number of words).
.TP
.I corpus\fR file containing the training corpus.
.TP 
.I wordmodel\fR target\-language word co\-occurrence model (previously trained by means of the \fB\-\-trainwrd\fR option).
.TP
.I dic\fR the lexical-selection dictionary (binary format).
.TP
.I bildic\fR the bilingual dictionary (binary format).
.TP
.I model\fR output file on which the co\-occurrence models are saved.
.PP
.PP
.B \-\-lextor | \-l
.br
Perform the lexical selection on the input stream. It needs the
following required parameters:
.TP
.I model\fR  file containing the model to be used for the lexical selection.
.TP
.I dic\fR lexical\-selection dictionary (binary format).
.TP
.I left\fR left\-side context to take into account (number of words).
.TP
.I right\fR right\-side context to take into account (number of words).
.PP
.B \-\-weightexp w
.br 
Specify a weight value to change the influence of surrounding words
while training or performing the lexical selection. The parameter
\fIw\fR must be a positive value.
.PP
.B \-\-debug | \-d
.br 
Show debug information while working.
.PP
.B \-\-help | \-h
.br
Shows this help.
.PP
.B \-\-version | \-v
.br
Shows license information.
.PP
.SH SEE ALSO
.I apertium\-gen\-lextorbil\fR(1),
.I apertium\-preprocess\-corpus\-lextor\fR(1),
.I apertium\-gen\-stopwords\-lextor\fR(1),
.I apertium\-gen\-wlist\-lextor\fR(1),
.I apertium\-gen\-wlist\-lextor\-translation\fR(1),
.I apertium\-lextor\-eval\fR(1),
.I apertium\-lextor\-mono\fR(1).
.SH BUGS
Lots of...lurking in the dark and waiting for you!
.SH AUTHOR
(c) 2005,2006 Universitat d'Alacant / Universidad de Alicante. All rights
reserved.