nb 0.0.1 → 0.0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 8c47b77ea4e1ea80b323d2d398a0fcaa8344b1d1
-  data.tar.gz: eecba79751ad4b93e8aeec5bd1b769a4aec40a29
+  metadata.gz: e163cd57a340190efef6d152493f3cbf5f0ee6c7
+  data.tar.gz: 3dc32f2eaf69f8cda5fe9672bdd54c4c51d47a57
 SHA512:
-  metadata.gz: 14fd940f7e3e7f99d426fc31a3de91f668ac9622df3aca7739c4d4f3c1c618aa9e63f6e5ab1fc994fbe7b6c71e226c3e50cb8725f0ee953e1675cf4047aa64d1
-  data.tar.gz: 34644c2f74430a5753f9f196906162430ff20faecada43db0c2f832108459aaa71337c6b5e267aa47a1d664a5290205c75841f3acf828365c2e22d600974ddc5
+  metadata.gz: fb8c29aa398fdda618c7c624146435a5d38386f4698f4a4262c5178826089e50568b5eb213af53f7f1de48d5e73f93e5fad9cbf47868403f4a522350f03c1b73
+  data.tar.gz: 3e3bc9a7f6371ad037fb62a1cf3a5cbeb747aa958bea96c8a14879174a3bfd754a7580612961916f6be926b9cc56a80be28add65572e8ad376ce274caf1a1fa5

data/.travis.yml

@@ -0,0 +1,5 @@
+language: ruby
+rvm:
+  - 2.0.0
+script: bundle exec rspec
+gemfile: Gemfile.ci

data/Gemfile.ci

@@ -0,0 +1,11 @@
+source 'https://rubygems.org'
+
+gemspec
+
+group :development, :test do
+  gem 'rspec', '~> 2.12.0'
+  gem 'guard'
+  gem 'guard-rspec'
+  gem 'simplecov', require: false
+  gem 'coveralls', require: false
+end

data/README.md

@@ -1,6 +1,9 @@
-# Nb
+# nb
 
-TODO: Write a gem description
+[![Code Climate](https://codeclimate.com/github/forresty/nb/badges/gpa.svg)](https://codeclimate.com/github/forresty/nb)
+[![Build Status](https://travis-ci.org/forresty/nb.svg?branch=master)](https://travis-ci.org/forresty/nb)
+
+yet another Naive Bayes library
 
 ## Installation
 
@@ -20,7 +23,42 @@ Or install it yourself as:
 
 ## Usage
 
-TODO: Write usage instructions here
+```ruby
+bayes = NaiveBayes.new :love, :hate
+
+bayes.train :love, 'I', 'love', 'you'
+bayes.train :hate, 'I', 'hate', 'you'
+
+bayes.classifications(*%w{ I love you }).should == [[:love, 0.5], [:hate, 0.25]]
+bayes.classify(*%w{ I love you }).should == [:love, 0.5]
+bayes.classify(*%w{ love }).should == [:love, 0.5]
+```
+
+### ability to view top tokens
+
+`bayes.top_tokens_of_category(:spam)`
+
+```
++------------+------+--------------------+
+| 学生       | 1966 | 0.9995149465854383 |
+| 多劳多得   | 1953 | 0.999511719439795  |
+| 党         | 1517 | 0.9993714712416684 |
+| 结         | 1327 | 0.9992815430836995 |
+| 工资       | 1213 | 0.9992140742313297 |
+| 不等       | 1135 | 0.999160108836817  |
+| 诚聘       | 1107 | 0.9991388832706672 |
+| 咨询       | 1095 | 0.9991294545902496 |
+| 加入       | 1071 | 0.9991099639327047 |
+| 限制       | 1046 | 0.9990887109454397 |
+| 50         | 1041 | 0.9990843379645474 |
+| 上网       | 1020 | 0.9990655037161098 |
+| 流动资金   | 952  | 0.9989988208099915 |
+| 曰         | 902  | 0.9989433817121107 |
+| 办公室     | 861  | 0.9988931222482719 |
+| 职员       | 827  | 0.9988476682254364 |
+| 绝对       | 823  | 0.9988420740701035 |
++------------+------+--------------------+
+```
 
 ## Contributing
 

data/lib/nb/naive_bayes.rb

@@ -33,6 +33,14 @@ class NaiveBayes
     scores.sort_by { |k, v| -v }
   end
 
+  def top_tokens_of_category(category, count=20)
+    tokens_count[category].map { |k, v| [k, v, probability_of_a_token_in_category(k, category)] }.sort_by { |i| -i.last }.first(count)
+  end
+
+  def probability_of_a_token_in_category(token, category)
+    probability_of_a_token_given_a_category(token, category) / @categories.inject(0.0) { |r, c| r + probability_of_a_token_given_a_category(token, c) }
+  end
+
   def probability_of_a_token_given_a_category(token, category)
     return assumed_probability if @tokens_count[category][token] == 0
 

data/lib/nb/version.rb

@@ -1,3 +1,3 @@
 class NaiveBayes
-  VERSION = "0.0.1"
+  VERSION = "0.0.2"
 end

data/spec/nb/naive_bayes_spec.rb

@@ -8,8 +8,10 @@ describe NaiveBayes do
   it { should respond_to :probability_of_a_token_given_a_category }
   it { should respond_to :probability_of_tokens_given_a_category }
   it { should respond_to :probability_of_a_category }
+  it { should respond_to :probability_of_a_token_in_category }
   # it { should respond_to :total_number_of_tokens }
   it { should respond_to :total_number_of_items }
+  it { should respond_to :top_tokens_of_category }
 
   let(:bayes) { NaiveBayes.new(:love, :hate) }
   subject { bayes }
@@ -27,6 +29,24 @@ describe NaiveBayes do
   #   end
   # end
 
+  describe '#probability_of_a_token_in_category' do
+    it 'calculates correctly' do
+      bayes.train :love, 'I', 'love', 'you'
+      bayes.train :hate, 'I', 'hate', 'you'
+
+      bayes.probability_of_a_token_in_category('love', :love).should == 2.0/3  # 1 / ( 1 + 0.5 )
+      bayes.probability_of_a_token_in_category('hate', :love).should == 1.0/3  # 0.5 / ( 1 + 0.5 )
+      bayes.probability_of_a_token_in_category('I', :love).should == 0.5
+
+      bayes.train :love, 'hate', 'is', 'love'
+      bayes.train :love, 'hate', 'is', 'love'
+      bayes.train :love, 'hate', 'is', 'love'
+
+      bayes.probability_of_a_token_in_category('love', :love).should == 5.0/6  # 1 / ( 1 + 0.2 )
+      bayes.probability_of_a_token_in_category('hate', :love).should == 3.0/7  # 0.75 / ( 0.75 + 1 )
+    end
+  end
+
   describe '#total_number_of_items' do
     it 'calculates correctly' do
       bayes.train :love, 'I', 'love', 'you'

data/spec/spec_helper.rb

@@ -3,4 +3,10 @@ SimpleCov.start do
   add_filter '/spec/'
 end
 
+begin
+  require 'coveralls'
+  Coveralls.wear!
+rescue LoadError
+end
+
 require "nb"

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: nb
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.0.2
 platform: ruby
 authors:
 - Forrest Ye
@@ -46,7 +46,9 @@ extensions: []
 extra_rdoc_files: []
 files:
 - .gitignore
+- .travis.yml
 - Gemfile
+- Gemfile.ci
 - Guardfile
 - LICENSE.txt
 - README.md