summa 0.0.9 → 0.0.10

data/Manifest.txt

@@ -4,9 +4,10 @@ PostInstall.txt
 README.rdoc
 Rakefile
 lib/summa.rb
-lib/FrequencyAnalyzer.rb
 lib/SummaUtils.rb
 lib/SummaData.rb
+lib/DocumentGraph.rb
+lib/FrequencyAnalyzer.rb
 script/console
 script/destroy
 script/generate

data/lib/DocumentGraph.rb

@@ -0,0 +1,94 @@
+require 'rubygems'
+require 'rgl/adjacency'
+require 'rgl/dot'
+require 'Matrix'
+
+class CGDocumentGraph
+	
+  def initialize(document, stopWordsDoc)
+    @document = document
+    @stopWords = stopWordsDoc
+    @adjacencyGraph = RGL::DirectedAdjacencyGraph.new()
+    @wordArray = []
+		
+    #Write the non-stop words into a word array.
+    for i in 0 ... @document.docArray.length() - 1
+      word = @document.docArray[i]
+      if word != nil
+        word = CGWordOps.removePunctuation(word)
+        if !@stopWords.docArray.include?(word.downcase)
+          @wordArray << word
+        end
+      end
+    end
+		
+    for i in 0 ... @wordArray.length() - 2
+      word1 = @wordArray[i]
+      word2 = @wordArray[i+1]
+			
+      @adjacencyGraph.add_edge(word1, word2)
+    end
+  end
+	
+  def buildKeywords(delta, thresh)
+    verts = @adjacencyGraph.vertices()
+    num_verts = @adjacencyGraph.size()
+    q_array = [];
+    for i in 0 .. (num_verts - 1)
+      w = verts[i]
+      n = @adjacencyGraph.out_degree(w)
+      if(n != 0)
+        q = Array.new(num_verts,0)
+        adjVerts = @adjacencyGraph.adjacent_vertices(w)
+        for j in 0 ... adjVerts.size() - 1
+          v = adjVerts[j]
+          index = verts.index(v)
+          if (index != nil)
+            q[index] = 1.0/n
+          end
+        end
+        q_array << q
+      else
+        q = Array.new(num_verts, 1.0/num_verts)
+        q_array << q
+      end
+    end
+
+    prM = Matrix.rows(q_array)
+		
+    prArray = Array.new(num_verts, 1.0/num_verts);
+    pr = Matrix.columns([prArray])
+		
+    deltaArray = Array.new(num_verts, (1.0-delta)/num_verts)
+    deltaM = Matrix.columns([deltaArray])
+		
+    for i in 0 .. 10
+      pr = deltaM + delta * prM * pr
+    end
+		
+    prArray = (pr.column_vectors[0].to_a)
+		
+    prHash = Hash.new()
+    for i in 0 .. num_verts - 1
+      prHash[prArray[i]] = verts[i]
+    end
+		
+    sortedHash = prHash.sort()
+    sortedHash = sortedHash.reverse()
+		
+    if(thresh >= num_verts)
+      thresh = num_verts - 1
+    end
+    keywords = Array.new()
+		
+    for i in 0 .. thresh
+      keywords << (sortedHash[i])[1]
+    end
+		
+    keywords
+  end
+	
+  attr_accessor :adjacencyGraph
+
+end
+

data/lib/SummaUtils.rb

@@ -1,5 +1,10 @@
 class String
-  def freqkeys
+
+  def removePunctuation!
+    self.delete!(".,-:;()?!\"\'*_")
+  end
+
+  def frequent
     @freqCount = {}
     @stopwords = SummaData.stopwords
     @mean = 0

data/lib/summa.rb

@@ -7,16 +7,12 @@ require 'SummaData'
 require 'SummaUtils'
 
 module Summa
-  VERSION = '0.0.9'
+  VERSION = '0.0.10'
 end
 
 class String
-  def summarize
-    puts "Testing version 0.0.9 "
-  end
-
-  def removePunctuation!
-    self.delete!(".,-:;()?!\"\'")
+  def summary
+		puts "10!" 
   end
 end
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: summa
 version: !ruby/object:Gem::Version 
-  version: 0.0.9
+  version: 0.0.10
 platform: ruby
 authors: 
 - HyLiter.org
@@ -42,9 +42,10 @@ files:
 - README.rdoc
 - Rakefile
 - lib/summa.rb
-- lib/FrequencyAnalyzer.rb
 - lib/SummaUtils.rb
 - lib/SummaData.rb
+- lib/DocumentGraph.rb
+- lib/FrequencyAnalyzer.rb
 - script/console
 - script/destroy
 - script/generate