nabatheon 0.0.2 → 0.0.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 4fa252b6b9ee408a134f7453076b1e17d2430a16
-  data.tar.gz: 31c076ac8c39735b8c6450ce961aea4a9ad16420
+  metadata.gz: 35fc21082f3b4056a1995d300683c37459743a5b
+  data.tar.gz: 5333c759976de623a600bd335588bb10eba7aaa1
 SHA512:
-  metadata.gz: 611dc73388d6adbc0054e17e70df524efceff6ae3b11436f4221cd87bc9fc4776e0634e51dace1b3124e379e062454090ea6e6f30324c2b2dd3f81ccd9b94678
-  data.tar.gz: d2ece7d10fdb835c080af46f3c6750968eb37f1a7e98417dc49df1c13edced95a651efec16f3e768e5d601224183aeaaf926b46888b73a5a047e87a1173bfecd
+  metadata.gz: 86d96b766265bdb24a105d7c25491f4a39d5313bbd32385abc2d05f2b3e98ad284b189c30e9ce967950321aeb6ba03b1e575b07cac64c6ce61798217015d6ba1
+  data.tar.gz: 52e74491a71c41a991610bbd4ac737469b2974c8e24d308b63b9a37a6987f9c9670cc68793629343d88ae36d226e664a50e9fa37cf123f94d01daff98ecb11fa

data/lib/nabatheon.rb

@@ -1,4 +1,8 @@
 require 'nabatheon/version'
+require 'nabatheon/entity'
+require 'nabatheon/rule'
+require 'nabatheon/search'
+
 require 'stanford-core-nlp'
 
 module Nabatheon
@@ -9,18 +13,22 @@ module Nabatheon
   StanfordCoreNLP.use :english
   Pipeline = StanfordCoreNLP.load(:tokenize, :ssplit, :pos, :lemma, :parse, :ner)
 
-  def self.annotate(raw_text)
-    text = StanfordCoreNLP::Annotation.new(raw_text)
-    Pipeline.annotate(text)
+  def self.annotate(text)
+    annotated_text = StanfordCoreNLP::Annotation.new(text)
+    Pipeline.annotate(annotated_text)
 
-    named_lemma = []
-    text.get(:sentences).each do |sentence|
+    tagged = []
+    annotated_text.get(:sentences).each do |sentence|
       sentence.get(:tokens).each do |token|
         entity_tag = token.get(:named_entity_tag).to_s
         lemma      = token.get(:lemma).to_s
-        named_lemma << [entity_tag, lemma] unless entity_tag == 'O'
+        tagged << { named_entity: entity_tag, type: lemma }
       end
     end
-    named_lemma
+    tagged
+  end
+
+  def self.relevant_searches_for(annotations)
+    Rule.apply_on(annotations)
   end
 end

data/lib/nabatheon/entity.rb

@@ -0,0 +1,3 @@
+module Nabatheon
+  Entity = Struct.new(:type, :value)
+end

data/lib/nabatheon/rule.rb

@@ -0,0 +1,34 @@
+module Nabatheon
+  class Rule
+    PERSON = 'PERSON'
+    ORG    = 'ORGANIZATION'
+
+    def self.apply_on(*args)
+      self.new.apply_on(*args)
+    end
+
+    def apply_on(named_entities)
+      searches = []
+      previous = Entity.new(nil, nil)
+      named_entities.map do |entity_arr|
+        entity = Entity.new(entity_arr[:named_entity], entity_arr[:type])
+        if rule_of_two?(entity.type, previous.type)
+          searches.last.value = "#{searches.last.value} #{entity.value}"
+        else
+          searches << Search.new(entity.type, entity.value)
+        end
+        previous = entity
+      end
+      searches
+    end
+
+    # * (2 persons || 2 orgs) in a row => #contact
+    def rule_of_two?(type, previous_type)
+      # FIXME: Does not return correct value if if the types are:
+      #        type = PERSON && previous[:type] = ORG
+      #        # => returns true
+      (type          == PERSON || type          == ORG) && \
+      (previous_type == PERSON || previous_type == ORG)
+    end
+  end
+end

data/lib/nabatheon/search.rb

@@ -0,0 +1,10 @@
+module Nabatheon
+  class Search
+    attr_accessor :type, :value
+
+    def initialize(type, value)
+      @type  = type
+      @value = value
+    end
+  end
+end

data/lib/nabatheon/version.rb

@@ -1,3 +1,3 @@
 module Nabatheon
-  VERSION = '0.0.2'
+  VERSION = '0.0.3'
 end

data/nabatheon.gemspec

@@ -8,7 +8,7 @@ Gem::Specification.new do |spec|
   spec.version       = Nabatheon::VERSION
   spec.authors       = ['Jacob Burenstam']
   spec.email         = ['burenstam@gmail.com']
-  spec.summary       = %q{Extract named entities and lemmas using StanfordCoreNLP.}
+  spec.summary       = %q{Extract named entities and lemmas using StanfordCoreNLP (alpha).}
   spec.description   = %q{Extract named entities and lemmas using StanfordCoreNLP (alpha). Assumes that all dependencies are in ~/.stanford-nlp}
   spec.homepage      = 'https://github.com/buren/nabatheon'
   spec.license       = 'MIT'
@@ -22,5 +22,4 @@ Gem::Specification.new do |spec|
 
   spec.add_development_dependency 'bundler', '~> 1.7'
   spec.add_development_dependency 'rake', '~> 10.0'
-  spec.add_development_dependency 'pry'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: nabatheon
 version: !ruby/object:Gem::Version
-  version: 0.0.2
+  version: 0.0.3
 platform: ruby
 authors:
 - Jacob Burenstam
@@ -52,20 +52,6 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '10.0'
-- !ruby/object:Gem::Dependency
-  name: pry
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
 description: Extract named entities and lemmas using StanfordCoreNLP (alpha). Assumes
   that all dependencies are in ~/.stanford-nlp
 email:
@@ -82,6 +68,9 @@ files:
 - Rakefile
 - bin/nabatheon
 - lib/nabatheon.rb
+- lib/nabatheon/entity.rb
+- lib/nabatheon/rule.rb
+- lib/nabatheon/search.rb
 - lib/nabatheon/version.rb
 - nabatheon.gemspec
 homepage: https://github.com/buren/nabatheon
@@ -107,6 +96,6 @@ rubyforge_project:
 rubygems_version: 2.2.2
 signing_key: 
 specification_version: 4
-summary: Extract named entities and lemmas using StanfordCoreNLP.
+summary: Extract named entities and lemmas using StanfordCoreNLP (alpha).
 test_files: []
 has_rdoc: