| Statistics-Suggest documentation | Contained in the Statistics-Suggest distribution. |
Statistics::Suggest - Perl binding for collaborative filtering library SUGGEST.
Download SUGGEST from http://glaros.dtc.umn.edu/gkhome/suggest/download.
Find libsuggest.a which matches your environment and place it under
your library path (or specify its path with LIBS option as shown below).
Then do:
perl Makefile.PL [LIBS='-L/where/to/find/libsuggest.a -lsuggest'] make make test make install
Tested with suggest-1.0-linux.
use Statistics::Suggest;
## initialize SUGGEST with $data
my $data = [
# array of [$user_id, $item_id], ...
[1, 1], [1, 2], [1, 4], [1, 5]
[2, 1], [2, 2], [2, 4],
[3, 3], [3, 4],
[4, 3], [4, 4], [4, 5],
...
];
my $s = new Statistics::Suggest(
RType => 2,
NNbr => 40,
Alpha => 0.3,
);
$s->load_trans($data);
$s->init;
## make top 10 recommendations for $selected_item_ids
my $rcmds;
my $selected_item_ids = [1, 2];
$s->top_n($selected_item_ids, 10, \$rcmds)
print "recommendations: " . join(',', @$rcmds);
This is a perl binding for SUGGEST. Please refer to the SUGGEST's manual for details. Basically, this package contains all corresponding methods for functions described in the manual.
my $s = new Statistics::Suggest(
# parameters for Init function (see SUGGEST's manual p.5)
RType => 2,
NNbr => 40,
Alpha => 0.3,
);
Loads user-item transactions that will be used to make recommendations.
$s->load_trans([
# [user_id, item_id], ...
[1, 1], [1, 2], [1, 4],...
]);
Alpha parameter can be set automatically by calling estimate_alpha method
after load_trans, before calling init.
$s->estimate_alpha;
Initializes the SUGGEST engine. Should be called after all transactions are loaded
via load_trans method.
$s->init;
Returns top-n recommendations for the given item set. Should be called after init.
my $rcmds;
$s->top_n(
[3, 4, 5, ... ], # array of item_ids in the user's basket
10, # number of recommendations required
\$rcmds, # reference of an array reference for storing recommendations
);
print join(',', @$rcmds);
None by default.
http://glaros.dtc.umn.edu/gkhome/suggest/overview
Ikuhiro IHARA <tsukue@gmail.com>
Copyright (C) 2007 by Ikuhiro IHARA
This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself, either Perl version 5.8.5 or, at your option, any later version of Perl 5 you may have available.
| Statistics-Suggest documentation | Contained in the Statistics-Suggest distribution. |
package Statistics::Suggest; #use 5.008008; use strict; use warnings; use Carp; require Exporter; use AutoLoader; our @ISA = qw(Exporter); # Items to export into callers namespace by default. Note: do not export # names by default without a very good reason. Use EXPORT_OK instead. # Do not simply export all your public functions/methods/constants. # This allows declaration use Statistics::Suggest ':all'; # If you do not need this, moving things directly into @EXPORT or @EXPORT_OK # will save memory. our %EXPORT_TAGS = ( 'all' => [ qw( ) ] ); our @EXPORT_OK = ( @{ $EXPORT_TAGS{'all'} } ); our @EXPORT = qw( ); our $VERSION = '0.01'; sub AUTOLOAD { # This AUTOLOAD is used to 'autoload' constants from the constant() # XS function. my $constname; our $AUTOLOAD; ($constname = $AUTOLOAD) =~ s/.*:://; croak "&Statistics::Suggest::constant not defined" if $constname eq 'constant'; my ($error, $val) = constant($constname); if ($error) { croak $error; } { no strict 'refs'; # Fixed between 5.005_53 and 5.005_61 #XXX if ($] >= 5.00561) { #XXX *$AUTOLOAD = sub () { $val }; #XXX } #XXX else { *$AUTOLOAD = sub { $val }; #XXX } } goto &$AUTOLOAD; } require XSLoader; XSLoader::load('Statistics::Suggest', $VERSION); # Preloaded methods go here. # Autoload methods go after =cut, and are processed by the autosplit program. sub new { my $class = shift; my $self = bless { RType => 2, NNbr => 20, Alpha => 0.4, @_ }, $class; return $self; } sub load_trans { my ($self, $trans) = @_; my $nusers = 0; my $nitems = 0; my $ntrans = 0; my @userid; my @itemid; for (@$trans) { my ($u, $i) = @$_; $ntrans ++; $nusers = $u if $u > $nusers; $nitems = $i if $i > $nitems; push @userid, $u; push @itemid, $i; } $self->{nusers} = $nusers; $self->{nitems} = $nitems; $self->{ntrans} = $ntrans; $self->{userid} = \@userid; $self->{itemid} = \@itemid; } sub init { my $self = shift; croak "necessary params not set" unless ( defined($self->{nusers}) and defined($self->{nitems}) and defined($self->{ntrans}) and defined($self->{userid}) and defined($self->{itemid}) and defined($self->{RType}) and defined($self->{NNbr}) and ($self->{NNbr} != 2 or defined($self->{Alpha})) ); $self->{RcmdHandle} = _SUGGEST_Init( map $self->{$_}, qw(nusers nitems ntrans userid itemid RType NNbr Alpha) ); } sub estimate_alpha { my ($self, $nrcmd) = @_; croak "necessary params not set" unless ( defined($self->{nusers}) and defined($self->{nitems}) and defined($self->{ntrans}) and defined($self->{RType}) and defined($self->{NNbr}) ); $self->{NRcmd} = ($nrcmd || $self->{NRcmd}); $self->{Alpha} = _SUGGEST_EstimateAlpha( map $self->{$_}, qw(nusers nitems ntrans userid itemid NNbr NRcmd) ); } sub top_n { my ($self, $itemids, $nrcmd, $rcmds) = @_; croak "should init first" unless $self->{RcmdHandle}; $self->{NRcmd} = ($nrcmd || $self->{NRcmd}); return _SUGGEST_TopN( $self->{RcmdHandle}, scalar @$itemids, $itemids, $self->{NRcmd}, $$rcmds ); } sub DESTROY { my $self = shift; if ($self->{RcmdHandle}) { _SUGGEST_Clean($self->{RcmdHandle}); } } 1; __END__ # Below is stub documentation for your module. You'd better edit it!