Help language development. Donate to The Perl Foundation

Algorithm::LDA cpan:TITSUKI last updated on 2019-01-15

Algorithm-LDA-0.0.9/

Build Status

NAME

Algorithm::LDA - A Perl 6 Latent Dirichlet Allocation implementation.

SYNOPSIS

EXAMPLE 1

use Algorithm::LDA;
use Algorithm::LDA::Formatter;
use Algorithm::LDA::LDAModel;

my @documents = (
    "a b c",
    "d e f",
);
my ($documents, $vocabs) = Algorithm::LDA::Formatter.from-plain(@documents);
my Algorithm::LDA $lda .= new(:$documents, :$vocabs);
my Algorithm::LDA::LDAModel $model = $lda.fit(:num-topics(3), :num-iterations(500));

$model.topic-word-matrix.say; # show topic-word matrix
$model.document-topic-matrix; # show document-topic matrix
$model.log-likelihood.say; # show likelihood 
$model.nbest-words-per-topic.say # show nbest words per topic

EXAMPLE 2

use Algorithm::LDA;
use Algorithm::LDA::Formatter;
use Algorithm::LDA::LDAModel;

# Note: You can get AP corpus as follows:
# $ wget "https://github.com/Blei-Lab/lda-c/blob/master/example/ap.tgz?raw=true" -O ap.tgz
# $ tar xvzf ap.tgz

my @vocabs = "./ap/vocab.txt".IO.lines;
my @documents = "./ap/ap.dat".IO.lines;
my $documents  = Algorithm::LDA::Formatter.from-libsvm(@documents);

my Algorithm::LDA $lda .= new(:$documents, :@vocabs);
my Algorithm::LDA::LDAModel $model = $lda.fit(:num-topics(20), :num-iterations(500));

$model.topic-word-matrix.say; # show topic-word matrix
$model.document-topic-matrix; # show document-topic matrix
$model.log-likelihood.say; # show likelihood 
$model.nbest-words-per-topic.say # show nbest words per topic

DESCRIPTION

Algorithm::LDA is a Perl 6 Latent Dirichlet Allocation implementation.

CONSTRUCTOR

new

Defined as:

submethod BUILD(:$!documents!, :$!vocabs! is raw) { }

Constructs a new Algorithm::LDA instance.

METHODS

fit

Defined as:

method fit(Int :$num-iterations = 500, Int :$num-topics!, Num :$alpha = 0.1e0, Num :$beta = 0.1e0, Int :$seed --> Algorithm::LDA::LDAModel)

Returns an Algorithm::LDA::LDAModel instance.

AUTHOR

titsuki titsuki@cpan.org

COPYRIGHT AND LICENSE

Copyright 2018 titsuki

This library is free software; you can redistribute it and/or modify it under the Artistic License 2.0.

The algorithm is from: