tagmemics 0.0.0.beta → 0.0.2

data/config/articles.txt

@@ -0,0 +1,3 @@
+a
+an
+the

data/config/conjunctions.txt

@@ -0,0 +1,7 @@
+for
+and
+nor
+but
+or
+yet
+so

data/config/linking_verbs.txt

@@ -0,0 +1,28 @@
+Be
+Am
+Is
+Are
+Was
+Were
+Has become
+Could have come
+Shall be
+Shall have been
+Have appeared
+Should have appeared
+Will be
+Will have been
+Had seemed
+Should have been
+Has been
+Have been
+Had been
+Can be
+May be
+Might be
+Should be
+Could be
+Become
+Would be
+Appear
+Seem

data/config/prepositions.txt

@@ -0,0 +1,202 @@
+'gainst
+'mongst
+'neath
+'twixt
+abaft
+abeam
+aboard
+about
+above
+absent
+according to
+across
+afore
+after
+against
+ago
+ahead of
+along
+alongside
+amid
+amidst
+among
+amongst
+anenst
+anent
+anti
+apart
+apart from
+apropos
+apud
+around
+as
+as far as
+as for
+as of
+as opposed to
+as per
+as regards
+as soon as
+as well as
+aside
+aside from
+astern of
+astride
+at
+at the behest of
+athwart
+atop
+away
+ayond
+ayont
+back to
+barring
+because of
+before
+behind
+behither
+below
+beneath
+beside
+besides
+between
+betwixen
+betwixt
+beyond
+biforn
+but
+by
+by means of
+by virtue of
+chez
+circa
+close to
+concerning
+considering
+contra
+cum
+despite
+down
+due to
+during
+ere
+except
+except for
+excluding
+failing
+far from
+following
+for
+for the sake of
+forby
+forenenst
+fornenst
+fornent
+from
+fromward
+froward
+frowards
+gainst
+given
+hence
+in
+in accordance with
+in addition to
+in case of
+in front of
+in lieu of
+in order to
+in place of
+in point of
+in re
+in spite of
+in to
+including
+inside
+inside of
+inside out
+instead of
+into
+left of
+like
+mid
+midst
+minus
+modulo
+near
+near to
+neath
+next
+next to
+notwithstanding
+o'
+of
+off
+on
+on account of
+on behalf of
+on to
+on top of
+onto
+opposite
+opposite of
+opposite to
+out
+out from
+out of
+outside
+outside of
+outwith
+over
+overthwart
+owing to
+pace
+past
+per
+plus
+prior to
+pro
+pursuant to
+qua
+rather than
+re
+regarding
+regardless of
+right of
+round
+sans
+save
+since
+subsequent to
+such as
+than
+thanks to
+that of
+through
+throughout
+till
+times
+to
+tofore
+toforn
+toward
+towards
+under
+underneath
+unlike
+until
+unto
+up
+up to
+upon
+versus
+via
+vice
+with
+with a view to
+with regard to
+with respect to
+withal
+within
+without
+worth

data/config/pronouns.txt

@@ -0,0 +1,53 @@
+I
+he
+her
+hers
+herself
+him
+himself
+his
+hisself
+it
+its
+itself
+me
+mine
+my
+myself
+one
+one's
+oneself
+our
+ours
+ourself
+ourselves
+she
+thee
+their
+theirs
+theirself
+theirselves
+them
+themself
+themselves
+they
+thine
+thou
+thy
+thyself
+us
+we
+who
+whom
+whomself
+whose
+whoself
+y'all
+ye
+you
+you all
+your
+yours
+yourself
+yourselves
+youse

data/lib/tagmemics.rb

@@ -1,35 +1,52 @@
 require_relative './tagmemics/word'
 require_relative './tagmemics/sentence'
+require_relative './tagmemics/load_data'
 
-
-module Lexicon
+module Tagmemics
   def self.parse(str)
-    ParsedSentence.new(str)
+    WordSet.new(str)
   end
-end
 
-# The output of Lexicon.parse
-class ParsedSentence
-  attr_accessor :nouns, :verbs, :articles, :adjectives, :adverbs,
-                :prepositions, :conjunctions, :pronouns
+  # The output of Tagmemics.parse
+  class WordSet
+    ARTICLES = %w(the an a)
+    CONJUNCTIONS = %w(for and nor but or yet so )
+    LINKING_VERBS = LoadData.contents_to_a('linking_verbs')
+    PRONOUNS = LoadData.contents_to_a('pronouns')
+    PREPOSITIONS = LoadData.contents_to_a('prepositions')
 
-  def initialize(str)
-    @str = str
-  end
+    attr_accessor :nouns, :verbs, :articles, :adjectives, :adverbs,
+                  :prepositions, :conjunctions, :pronouns, :collection
 
-  def sentence_to_array(sentence)
-    sentence.split(/\W/)
-  end
+    def initialize(str)
+      @collection = []
+      arr = WordSet.sentence_to_array(str)
+      arr.each { |word| @collection << Word.new(word) }
+      # @set = WordSet.start_hash(WordSet.sentence_to_array(str))
+    end
+
+    class << self
+      include LoadData
+
+      # Will probably want to use punctuation in the future.
+      # For now, this removes it.
+      def sentence_to_array(sentence)
+        sentence.split(/\s+|\W+\z/)
+      end
 
-  def start_hash(arr)
-    arr.map do |word|
-      result =
-        case
-        when part_of_speech(ARTICLES, word).any? then :article
-        when part_of_speech(CONJUNCTIONS, word).any? then :conjunction
-        when part_of_speech(PRONOUNS, word).any? then :pronoun
-        end
-      [word, result]
-    end.to_h
+      # Moved part of speech.  This will not work right now.
+      # Probably need to delete this.
+      def start_hash(arr)
+        arr.map do |word|
+          result =
+            case
+            when part_of_speech(ARTICLES, word).any? then :article
+            when part_of_speech(CONJUNCTIONS, word).any? then :conjunction
+            when part_of_speech(PRONOUNS, word).any? then :pronoun
+            end
+          [word, result]
+        end.to_h
+      end
+    end
   end
 end

data/lib/tagmemics/{config.rb → load_data.rb}

@@ -1,5 +1,5 @@
 # Retrieves data from config folder to save to constants.
-module Config
+module LoadData
   def self.config_path
     File.join(File.dirname(__FILE__), '../../config')
   end
@@ -24,7 +24,7 @@ module Config
     page = agent.get(uri)
     destination = "./config/#{part_of_speech}.txt"
     target = page.search(css_selector)
-    regx = /[^'a-zA-Z\s]/ # anything beside letters, apostrophe or space
+    regx = /[^\047a-zA-Z\s]/ # \047 is an apostrophe
 
     arr = []
     target.each do |x|

data/lib/tagmemics/version.rb

@@ -0,0 +1,3 @@
+module Tagmemics
+  VERSION = '0.0.2'
+end

data/lib/tagmemics/word.rb

@@ -1,69 +1,59 @@
 require 'wordnet'
 require 'facets'
-require_relative './config'
+require_relative './load_data'
 require_relative './word/wordnet'
+require_relative './word/confidence'
 
-module Lexicon
+module Tagmemics
   class Word
-    include Config
-
-    ARTICLES = %w(the an a)
-    CONJUNCTIONS = %w(for and nor but or yet so )
-    PRONOUNS = Config.contents_to_a('pronouns')
-
-
-    def part_of_speech(constant, str)
-      arr = []
-      constant.each do |word|
-        regx = /\b#{word}\b/i
-        arr << word if regx =~ str # word phrase matches
-      end
-      arr
-    end
-
-    def decimal_complete(hsh)
-      total = hsh.length
-      complete = hsh.count { |_k, v| v } # not nil
-      complete / total.to_f
-    end
-
     def initialize(word)
-      @word = word
-      @confidence_levels = confidence_levels(word)
-    end
-
-    def confidence_levels(word)
-      {
-        :noun => noun_confidence(word),
-        :verb => verb_confidence(word),
-        :adjective => adjective_confidence(word),
-        :adverb => adverb_confidence(word),
-        :article => article_confidence(word),
-        :preposition => preposition_confidence(word),
-        :conjunction => conjunction_confidence(word)
-      }
-    end
-
-    def noun_confidence(str)
-      (WordNet.orig_probability(str) / 1) * 3
-    end
-
-    def verb_confidence(str)
-    end
-
-    def adjective_confidence(str)
-    end
+      @str = word
+      puts "examining: #{word}"
+      @tagmemic_confidence = Word.confidence_levels(word)
+    end
+
+    class << self
+      ##
+      # Because WordNet only tracks verbs, nouns, adverbs and adjectives,
+      # confidence levels can only be updated for those values.  The other words
+      #  such as pronouns, prepositions, and conjunctions are based off of list
+      #  in config folder.  Their score is pass or fail and is
+      #  calculated as 0 or 1.0.
+      def confidence_levels(word)
+        word = word.downcase
+        known_hsh = determine_known_words(word)
+        wordnet_hsh = determine_wordnet_words(word)
+
+        hsh = everything_nil(known_hsh) ? wordnet_hsh : known_hsh
+        delete_nogos(hsh)
+      end
 
-    def adverb_confidence(str)
-    end
+      def determine_known_words(word)
+        {
+          :article => article_confidence(word),
+          :preposition => preposition_confidence(word),
+          :pronoun => pronoun_confidence(word),
+          :conjunction => conjunction_confidence(word),
+          :linking_verb => linking_verb_confidence(word)
+        }
+      end
 
-    def article_confidence(str)
-    end
+      def everything_nil(hsh)
+        (hsh.select { |_k, v| v != 0.0 && !v.nil? }.empty?)
+      end
 
-    def preposition_confidence(str)
-    end
+      def delete_nogos(hsh)
+        hsh.delete_if { |_k, v| v == 0.0 || v.nil? }
+      end
 
-    def conjunction_confidence(str)
+      def determine_wordnet_words(word)
+        {
+          :noun => WordNetMethods.wordnet_probability(word, 'noun'),
+          :verb => WordNetMethods.wordnet_probability(word, 'verb'),
+          :adjective => WordNetMethods.wordnet_probability(word, 'adjective'),
+          :adverb => WordNetMethods.wordnet_probability(word, 'adverb')
+        }
+      end
     end
   end
 end