ppe-dbpediafinder 0.0.1

data/README

@@ -0,0 +1,23 @@
+dbpediafinder by Yves Raimond
+
+A very small Ruby gems for finding URIs in DBpedia, given a label and an optional disambiguation string.
+
+Works in two steps:
+
+ 1) Uses Google AJAX API to find out possible candidates
+ 2) Uses DBpedia SPARQL end-point to validate one of those candidates
+
+Example use:
+
+  require 'rubygems'
+  require 'dbpediafinder'
+
+  finder = DbpediaFinder::Finder.new
+  finder.find('j.s. bach', 'composer')
+=> ["Johann Sebastian Bach", "http://dbpedia.org/resource/Johann_Sebastian_Bach"]
+
+  finder.find('david aaronovitch', 'bbc radio presenter')
+=> ["David Aaronovitch", "http://dbpedia.org/resource/David_Aaronovitch"]
+
+  finder.find('isle of wight')
+=> ["Isle of Wight", "http://dbpedia.org/resource/Isle_of_Wight"]

data/dbpediafinder.gemspec

@@ -0,0 +1,13 @@
+Gem::Specification.new do |s|
+  s.name = "ppe-dbpediafinder"
+  s.version = "0.0.1"
+  s.date = "2009-11-19"
+  s.summary = "DBpedia finder"
+  s.email = "yves@dbtune.org"
+  s.homepage = "http://github.com/moustaki/dbpediafinder"
+  s.description = "A small rubygem for dbpedia lookups"
+  s.has_rdoc = true
+  s.authors = ['Yves Raimond']
+  s.files = ["README", "dbpediafinder.gemspec", "lib", "lib/dbpediafinder.rb", "lib/dbpedia_finder/finder.rb"]
+end
+

data/lib/dbpedia_finder/finder.rb

@@ -0,0 +1,72 @@
+require 'rubygems'
+require '4store-ruby'
+require 'uri'
+require 'json'
+require 'net/http'
+
+module DbpediaFinder
+
+  class Finder
+
+    def initialize
+      @store = FourStore::Store.new 'http://dbpedia.org/sparql'
+      @proxy = URI.parse(ENV['HTTP_PROXY']) if ENV['HTTP_PROXY']
+    end
+
+    def find(label, disambiguation = nil)
+      results = google_search(label, disambiguation)
+      results.each do |uri|
+        dbpedia = wikipedia_to_dbpedia(uri)
+        next if dbpedia.split('Category:').size > 1
+        query = "
+          PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
+          SELECT DISTINCT ?label WHERE {
+            <#{dbpedia}> rdfs:label ?label
+            FILTER (
+              regex(?label, '#{clean_label(label)}', 'i')
+            )
+          }
+        "
+        match = @store.select query
+        return [match[0]['label'], dbpedia] if match.size > 0
+      end
+      return nil
+    end
+
+    def google_search(label, disambiguation)
+      if disambiguation
+        query = "\"#{label}\" #{disambiguation} site:en.wikipedia.org"
+      else
+        query = "\"#{label}\" site:en.wikipedia.org"
+      end
+      query = URI.encode(query)
+      google_url_s = "http://ajax.googleapis.com/ajax/services/search/web?v=1.0&q=#{query}"
+      url = URI.parse(google_url_s)
+      if @proxy
+        h = Net::HTTP::Proxy(@proxy.host, @proxy.port).new(url.host, url.port)
+      else
+        h = Net::HTTP.new(url.host, url.port)
+      end
+      h.start do |h|
+        res = h.get(url.path + "?" + url.query)
+        json = JSON.parse(res.body)
+        results = json["responseData"]["results"].map { |result| result["url"] }
+        return results
+      end
+    end
+
+    def wikipedia_to_dbpedia(wikipedia)
+      url_key = wikipedia.split('/').last
+      return "http://dbpedia.org/resource/" + url_key
+    end
+
+    def clean_label(label)
+      # Remove initials (as they are expanded in Wikipedia/DBpedia labels)
+      cleaned_label = label.split(' ').select { |l| l.split('.').size == 1 }.join(' ')
+      cleaned_label = cleaned_label.split("'").join("\\'")
+      cleaned_label
+    end
+
+  end
+
+end

data/lib/dbpediafinder.rb

@@ -0,0 +1 @@
+require File.join(File.expand_path(File.dirname(__FILE__)), 'dbpedia_finder', 'finder.rb')

metadata

@@ -0,0 +1,58 @@
+--- !ruby/object:Gem::Specification 
+name: ppe-dbpediafinder
+version: !ruby/object:Gem::Version 
+  version: 0.0.1
+platform: ruby
+authors: 
+- Yves Raimond
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2009-11-19 00:00:00 +00:00
+default_executable: 
+dependencies: []
+
+description: A small rubygem for dbpedia lookups
+email: yves@dbtune.org
+executables: []
+
+extensions: []
+
+extra_rdoc_files: []
+
+files: 
+- README
+- dbpediafinder.gemspec
+- lib/dbpediafinder.rb
+- lib/dbpedia_finder/finder.rb
+has_rdoc: true
+homepage: http://github.com/moustaki/dbpediafinder
+licenses: []
+
+post_install_message: 
+rdoc_options: []
+
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.3.5
+signing_key: 
+specification_version: 3
+summary: DBpedia finder
+test_files: []
+