Free relatedness software for linux

WordNet::Similarity is a collection of Perl modules for computing measures of semantic relatedness.

SYNOPSIS

Basic Usage Example

use WordNet::QueryData;

use WordNet::Similarity::path;

my $wn = WordNet::QueryData->new;

my $measure = WordNet::Similarity::path->new ($wn);

my $value = $measure->getRelatedness("car#n#1", "bus#n#2");

my ($error, $errorString) = $measure->getError();

die $errorString if $error;

print "car (sense 1) bus (sense 2) = $valuen";

Using a configuration file to initialize the measure

use WordNet::Similarity::path;

my $sim = WordNet::Similarity::path->new($wn, "mypath.cfg");

my $value = $sim->getRelatedness("dog#n#1", "cat#n#1");

($error, $errorString) = $sim->getError();

die $errorString if $error;

print "dog (sense 1) cat (sense 1) = $valuen";

Printing traces

print "Trace String -> ".($sim->getTraceString())."n";

Introduction

We observe that humans find it extremely easy to say if two words are related and if one word is more related to a given word than another. For example, if we come across two words, car and bicycle, we know they are related as both are means of transport. Also, we easily observe that bicycle is more related to car than fork is. But is there some way to assign a quantitative value to this relatedness? Some ideas have been put forth by researchers to quantify the concept of relatedness of words, with encouraging results.

Eight of these different measures of relatedness have been implemented in this software package. A simple edge counting measure and a random measure have also been provided. These measures rely heavily on the vast store of knowledge available in the online electronic dictionary -- WordNet. So, we use a Perl interface for WordNet called WordNet::QueryData to make it easier for us to access WordNet. The modules in this package REQUIRE that the WordNet::QueryData module be installed on the system before these modules are installed.

WebService::GoogleHack::Rate is a Perl module that implements a simple relatedness measure and semantic orientation related type functions.

SYNOPSIS

use WebService::GoogleHack::Rate;

#GIVE PATH TO INPUT FILE HERE

my $INPUTFILE="";

#GIVE PATH TO TRACE FILE HERE

my $TRACEFILE="";

#create an object of type Rate

my $rate = WebService::GoogleHack::Rate->new();

$results=$rate->measureSemanticRelatedness1("dog", "cat");

#The PMI measure is stored in the variable $results, and it can also
#be accessed as $rate->{PMI};

$results=$rate->predictSemanticOrientation($INPUTFILE, "excellent", "bad",$TRACEFILE);

#The resutls can be accessed through
print $results->{prediction}."n";
$results->{PMI Measure}."n";
$rate->{prediction} &."n";
$rate->{PMI Measure}."n";

WebService::GoogleHack::Rate - This package uses Google to do some basic natural language processing. For example, given two words, say "knife" and "cut", the module has the ability to retrieve a semantic relatedness measure, commonly known as the PMI (Pointwise mututal information) measure. The larger the measure the more related the words are. The package can also predict the semantic orientation of a given paragraph of english text. A positive measure means that the paragraph has a positive meaning, and negative measure means the opposite.

WebService::GoogleHack is a Perl package that ties together all GoogleHack modules (Webservice::GoogleHack::Search, Webservice::GoogleHack::Spelling, Webservice::GoogleHack::Rate, and Webservice::GoogleHack::Text) to implement Natural Language Processing techniques that use the World Wide Web as a source of information. Use this package to access all the functionality of GoogleHack.
SYNOPSIS
use WebService::GoogleHack;
my $google = new WebService::GoogleHack;
#Initializing the object to the contents of the configuration file
# API Key, GoogleSearch.wsdl file location.
$google->initConfig("initconfig.txt");
#Printing the contents of the configuration file
$google->printConfig();
#Measure the semantic relatedness between the words "white house" and
#"president".
$measure=$google->measureSemanticRelatedness1("white house","president");
print "nRelatedness measure between white house and president is: ";
print $measure."n";
#Going to search for words that are related to "toyota" and "ford"
my @terms=();
push(@terms,"toyota");
push(@terms,"ford");
#The parameters are the search terms, number of web page results to look
#at, the number of iterations,output file and the "true" indicates that the
#diagnostic data should be stored in the file "results.txt"
$results=$google->Algorithm1(@terms,10,25,1,"results.txt","true");
print $results;
WebService::GoogleHack is a PERL package that interacts with the Google API, and implements basic functions that allow the user to interact with Google and retrieve results in an easy to use format. GoogleHack also implements and extends a number of Natural Language Processing by using the World Wide Web as a source of information.
Main features:
- Issue queries to Google (WebService::GoogleHack, WebService::GoogleHack::Search)
- Retrieve Spelling suggestions from Google (WebService::GoogleHack, WebService::GoogleHack::Spelling)
- Find the Pointwise Mututal Information (PMI) measure between two words (WebService::GoogleHack,WebService::GoogleHack::Rate)
- Given a paragraph find if the paragraph has a positive or negative semantic orientation.(WebService::GoogleHack,WebService::GoogleHack::Rate)
- Given a set of words along with a positively oriented word such as "excellent" and a negatively oriented word such as "poor", find if the word has a positive or negative semantic orientation.(WebService::GoogleHack,WebService::GoogleHack::Rate)
- Given a set of phrases along with a positively oriented word such as "excellent" and a negatively oriented word such as "poor", predict if the given phrases are positive or negative in sentiment.(WebService::GoogleHack,WebService::GoogleHack::Rate)
- Given two or more words finds a set of related words. (WebService::GoogleHack)