bayes_classifier 0.0.1.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: d164296489c1f693f53f141b5233c8fc76babdcc
+  data.tar.gz: 65d421b448594b4e70c52b6841b4993c6276dcda
+SHA512:
+  metadata.gz: b39634094b910f7cca0822803e10a378382b57d061f988a226dc62bae1d8685298959b7d8a16dc1b4c0091f13fb15332503f3fc71ce700478cb5fafe2ab790af
+  data.tar.gz: 953add18915f3bfa1881efb6c1a99658096fcdaffa14f4215778e3211517210b228e73405ecd76882856d810de424f7453e68e18c2607393deedeb9bc477ab3c

data/.gitignore

@@ -0,0 +1,3 @@
+pkg
+spec/reports
+tmp

data/.rspec

@@ -0,0 +1,2 @@
+--color
+--format Fuubar

data/Gemfile

@@ -0,0 +1,4 @@
+source 'https://rubygems.org'
+
+# Specify your gem's dependencies in bayes_classifier.gemspec
+gemspec

data/Gemfile.lock

@@ -0,0 +1,34 @@
+PATH
+  remote: .
+  specs:
+    bayes_classifier (0.0.1.1)
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    diff-lcs (1.2.4)
+    fuubar (1.1.1)
+      rspec (~> 2.0)
+      rspec-instafail (~> 0.2.0)
+      ruby-progressbar (~> 1.0)
+    rake (10.1.0)
+    rspec (2.14.1)
+      rspec-core (~> 2.14.0)
+      rspec-expectations (~> 2.14.0)
+      rspec-mocks (~> 2.14.0)
+    rspec-core (2.14.5)
+    rspec-expectations (2.14.2)
+      diff-lcs (>= 1.1.3, < 2.0)
+    rspec-instafail (0.2.4)
+    rspec-mocks (2.14.3)
+    ruby-progressbar (1.2.0)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  bayes_classifier!
+  bundler (~> 1.3)
+  fuubar
+  rake
+  rspec

data/LICENSE.txt

@@ -0,0 +1,22 @@
+Copyright (c) 2013 DarthSim
+
+MIT License
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

data/README.md

@@ -0,0 +1,55 @@
+# Bayes::Classifier
+
+Bayes::Classifier allows you to classify strings with naive Bayes classifier.
+
+## Installation
+
+Just add the following line to your `Gemfile`:
+
+```ruby
+gem 'bayes_classifier'
+```
+
+Then run 'bundle install'.
+
+## Usage
+
+```ruby
+# Create new classifier
+classifier = Bayes::Classifier.new
+
+# Train classifier with a string
+classifier.train :category1, "lorem ipsum dolor sit amet"
+
+# Train classifier with array of strings
+classifier.train_with_array :category2, ["the first string", "the second string", "the third string"]
+
+# Train classifier with textfile
+classifier.train_with_file :category3, "data/category3.txt"
+
+# Train classifier with CSV file (first column - string, second column - category)
+classifier.train_with_csv "data/training.csv"
+
+# Apply weighting to the top words of category
+classifier.apply_weighting :category3, 10
+
+# Remove empty categories
+classifier.pop_unused
+
+# Classify string
+classifier.classify "the string"
+
+# Reset categories
+classifier.flush
+
+# Remove all categories
+classifier.flush_all
+```
+
+## Contributing
+
+1. Fork it
+2. Create your feature branch (git checkout -b my-new-feature)
+3. Commit your changes (git commit -am 'Add some feature')
+4. Push to the branch (git push origin my-new-feature)
+5. Create new Pull Request

data/Rakefile

@@ -0,0 +1 @@
+require "bundler/gem_tasks"

data/bayes_classifier.gemspec

@@ -0,0 +1,25 @@
+# coding: utf-8
+lib = File.expand_path('../lib', __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require 'bayes_classifier/version'
+
+Gem::Specification.new do |spec|
+  spec.name          = "bayes_classifier"
+  spec.version       = Bayes::VERSION
+  spec.authors       = ["DarthSim"]
+  spec.email         = ["darthsim@gmail.com"]
+  spec.description   = "Naive Bayes classifier"
+  spec.summary       = "Allows to classify strings with naive Bayes classifier"
+  spec.homepage      = "https://github.com/DarthSim/bayes_classifier"
+  spec.license       = "MIT"
+
+  spec.files         = `git ls-files`.split($/)
+  spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  spec.test_files    = spec.files.grep(%r{^spec/})
+  spec.require_paths = ["lib"]
+
+  spec.add_development_dependency "bundler", "~> 1.3"
+  spec.add_development_dependency "rake"
+  spec.add_development_dependency "rspec"
+  spec.add_development_dependency "fuubar"
+end

data/lib/bayes.rb

@@ -0,0 +1,3 @@
+require "bayes/string"
+require "bayes/classifier"
+require "bayes/category"

data/lib/bayes/category.rb

@@ -0,0 +1,67 @@
+module Bayes
+  class Category
+    MIN_SCORE = 0.0000001
+
+    def initialize
+      reset
+    end
+
+    def reset
+      @words = {}
+      @words_count = 0
+    end
+
+    def train(text)
+      text.word_hash.each do |word, count|
+        @words[word] = @words[word].to_i + count
+        @words_count += count
+      end
+    end
+
+    def forget(text)
+      text.word_hash.each do |word, count|
+        @words[word] = @words[word].to_i - count
+        @words.delete(word) if @words[word] == 0
+        @words_count -= count
+      end
+    end
+
+    def apply_weighting(coeff)
+      top_words.each do |word|
+        apply_weighting_for word, coeff
+      end
+    end
+
+    def apply_weighting_for(word, coeff)
+      if old_weight = @words[word]
+        @words[word] = old_weight * coeff
+        @words_count += @words[word] - old_weight
+      end
+    end
+
+    def top_words(num = 100)
+      @words.sort_by{ |w,c| -c }.slice(0,num).map{ |w| w[0] }
+    end
+
+    def score_for(words)
+      if @words_count > 0
+        words = words.word_hash.keys unless words.is_a? Array
+
+        if words.any?
+          words.map do |word|
+            word_value = @words[word] || MIN_SCORE
+            Math.log(word_value / @words_count.to_f)
+          end.inject(:+)
+        else
+          Math.log(MIN_SCORE / @words_count)
+        end
+      else
+        -Float::INFINITY
+      end
+    end
+
+    def blank?
+      @words_count == 0
+    end
+  end
+end

data/lib/bayes/classifier.rb

@@ -0,0 +1,55 @@
+module Bayes
+  class Classifier
+    attr_reader :categories
+
+    def initialize
+      @categories = {}
+    end
+
+    def train(category, text)
+      ensure_category(category).train(text)
+    end
+
+    def ensure_category(category)
+      @categories[category] ||= Bayes::Category.new
+    end
+
+    def train_with_array(category, lines)
+      lines.each{ |line| train(category, line) }
+    end
+
+    def train_with_file(category, filename)
+      train_with_array category, File.read(filename).split(/\r?\n/)
+    end
+
+    def train_with_csv(filename, separator: "||")
+      csv = CSV.new File.read(filename), col_sep: separator, quote_char: "§" # hope § won't be used anywhere
+      csv.each do |row|
+        train row[1], row[0]
+      end
+    end
+
+    def apply_weighting(category, coeff)
+      ensure_category(category).apply_weighting(coeff)
+    end
+
+    def classify(string)
+      words = string.word_hash.keys
+      @categories.each_with_object({}) do |category, hash|
+        hash[category[0]] = category[1].score_for(words)
+      end.sort_by { |cat| -cat[1] }[0][0]
+    end
+
+    def pop_unused
+      @categories.delete_if{ |name,cat| cat.blank? }
+    end
+
+    def flush
+      @categories.each{ |name, cat| cat.reset }
+    end
+
+    def flush_all
+      @categories = {}
+    end
+  end
+end

data/lib/bayes/string.rb

@@ -0,0 +1,107 @@
+class String
+
+  # Returns a Hash of words and their frequencies
+  def word_hash
+    split_words.each_with_object({}) do |word, hash|
+      word.downcase!
+      if !word.stopword? && word.length > 2
+        hash[word] ||= 0
+        hash[word] += 1
+      end
+    end
+  end
+
+  def split_words
+    gsub(/[^\w\s]+/," ").split
+  end
+
+  def stopword?
+    STOPWORDS.include? self
+  end
+
+  private
+
+  STOPWORDS = [
+    "a",
+    "again",
+    "all",
+    "along",
+    "are",
+    "also",
+    "an",
+    "and",
+    "as",
+    "at",
+    "but",
+    "by",
+    "came",
+    "can",
+    "cant",
+    "couldnt",
+    "did",
+    "didn",
+    "didnt",
+    "do",
+    "doesnt",
+    "dont",
+    "ever",
+    "first",
+    "from",
+    "have",
+    "her",
+    "here",
+    "him",
+    "how",
+    "i",
+    "if",
+    "in",
+    "into",
+    "is",
+    "isnt",
+    "it",
+    "itll",
+    "just",
+    "last",
+    "least",
+    "like",
+    "most",
+    "my",
+    "new",
+    "no",
+    "not",
+    "now",
+    "of",
+    "on",
+    "or",
+    "should",
+    "sinc",
+    "so",
+    "some",
+    "th",
+    "than",
+    "this",
+    "that",
+    "the",
+    "their",
+    "then",
+    "those",
+    "to",
+    "told",
+    "too",
+    "true",
+    "try",
+    "until",
+    "url",
+    "us",
+    "were",
+    "when",
+    "whether",
+    "while",
+    "with",
+    "within",
+    "yes",
+    "you",
+    "youll",
+    ].freeze
+
+end

data/lib/bayes/test.rb

@@ -0,0 +1,81 @@
+require "csv"
+
+module Bayes
+  module Stats
+
+    ### Error Analysis ====================================
+
+    def self.error_analysis(classifier, category, positive_items, negative_items)
+      true_positives = 0
+      true_negatives = 0
+      false_negatives = 0
+      false_positives = 0
+
+      positive_items.each do |i|
+        if classifier.classify(i) == category
+          true_positives += 1.0
+        else
+          false_negatives += 1.0
+        end
+      end
+
+      negative_items.each do |i|
+        if classifier.classify(i) == category
+          false_positives += 1.0
+        else
+          true_negatives += 1.0
+        end
+      end
+
+      precision = true_positives / (true_positives + false_positives)
+      recall = true_positives / (true_positives + false_negatives)
+      f_score = 2 * ( (precision * recall) / (precision + recall) )
+
+      {
+        true_positives: true_positives,
+        true_negatives: true_negatives,
+        false_negatives: false_negatives,
+        false_positives: false_positives,
+        precision: precision,
+        recall: recall,
+        f_score: f_score,
+      }
+    end
+
+    def self.error_analysis_csv(classifier, filename)
+      items = File.read(filename).split("\n").map {|t| t.split("||") }
+
+      correct = 0
+      incorrect = 0
+
+      items.each do |item|
+        category = classifier.classify(item.first)
+        if category == item.last
+          correct += 1
+        else
+          incorrect += 1
+        end
+      end
+
+      {
+        correct: correct,
+        incorrect: incorrect,
+        error_rate: incorrect / (incorrect + correct).to_f
+      }
+    end
+
+    ### Helpers ===================================================
+
+    def self.to_csv(results, name: "examples")
+      `mkdir -p spec/reports`
+
+      CSV.open("spec/reports/#{name}.csv", "w+") do |csv|
+        csv << results.first.keys
+        results.each do |r|
+          csv << r.values
+        end
+      end
+    end
+
+  end
+end