ruby client fix + documentation

clients/ruby/client/lib/relevanced_client.rb

@@ -52,7 +52,7 @@ def create_document(text)
         end
 
         def create_document_with_id(document_id, text)
-            @thrift_client.createDocumentWithId(
+            @thrift_client.createDocumentWithID(
                 document_id, text
             )
         end

docs/examples/ruby-binary-classifier.md

@@ -0,0 +1,73 @@
+## Binary Classifier (Ruby)
+
+In this example, each of a small collection of Wikipedia articles is added to a relevanced server and assigned to the centroid `math`.  Similarity scores for another math-related article and an irrelevant article on Richard Gere are then calculated.
+
+This example uses [nokogiri](http://www.nokogiri.org/) to crawl the Wikipedia pages and extract their text.  Note that `open-uri` is part of the Ruby standard library.
+
+Also note that by far the slowest part of the script is fetching and crawling the Wikipedia pages.
+
+```ruby
+require 'relevanced_client'
+require 'nokogiri'
+require 'open-uri'
+
+def run()
+    client = RelevancedClient::Client.new('localhost', 8097)
+    client.create_centroid('math')
+
+    MATH_URLS.each do |url|
+        text = fetch_url(url)
+        client.create_document_with_id(url, text)
+        client.add_document_to_centroid('math', url)
+    end
+
+    client.join_centroid('math')
+
+    TEST_URLS.each do |url|
+        text = fetch_url(url)
+        similarity = client.get_text_similarity('math', text)
+        puts "#{url} -> #{similarity}"
+    end
+end
+
+def fetch_url(url)
+    page = Nokogiri::HTML(open(url))
+    paragraphs = page.css('#mw-content-text > p').map { |x| x.text }
+    paragraphs.join("\n")
+end
+
+MATH_URLS = [
+    'https://en.wikipedia.org/wiki/Linear_programming',
+    'https://en.wikipedia.org/wiki/Algorithm',
+    'https://en.wikipedia.org/wiki/Scalar_multiplication',
+    'https://en.wikipedia.org/wiki/Mathematical_structure',
+    'https://en.wikipedia.org/wiki/Dot_product',
+    'https://en.wikipedia.org/wiki/Analysis_of_algorithms',
+    'https://en.wikipedia.org/wiki/Linear_algebra',
+    'https://en.wikipedia.org/wiki/Criss-cross_algorithm',
+    'https://en.wikipedia.org/wiki/Support_vector_machine',
+    'https://en.wikipedia.org/wiki/Mathematics',
+    'https://en.wikipedia.org/wiki/Rational_number',
+    'https://en.wikipedia.org/wiki/Fraction_(mathematics)',
+    'https://en.wikipedia.org/wiki/Square_root_of_2',
+    'https://en.wikipedia.org/wiki/Mathematical_optimization',
+    'https://en.wikipedia.org/wiki/Optimization_problem',
+    'https://en.wikipedia.org/wiki/Candidate_solution',
+    'https://en.wikipedia.org/wiki/Search_algorithm'
+]
+
+TEST_URLS = [
+    'https://en.wikipedia.org/wiki/Matrix_multiplication',
+    'https://en.wikipedia.org/wiki/Richard_Gere'
+]
+
+begin
+    run()
+end
+```
+
+As of 10/12/2015 (the article text may change), the result is:
+```bash
+https://en.wikipedia.org/wiki/Matrix_multiplication -> 0.4271200925196943
+https://en.wikipedia.org/wiki/Richard_Gere -> 0.07472419631570816
+```

mkdocs.yml

@@ -12,6 +12,7 @@ pages:
   - Binary Classifier (Python): examples/python-binary-classifier.md
   - Multiclass Classifier (Python): examples/python-multiclass-classifier.md
   - Binary Classifier (Javascript): examples/javascript-binary-classifier.md
+  - Binary Classifier (Ruby): examples/ruby-binary-classifier.md
   - Binary Classifier (Scala): examples/scala-binary-classifier.md
   - Binary Classifier (Java): examples/java-binary-classifier.md
   - Multiclass Classifier (Java): examples/java-multiclass-classifier.md