solid_naive_bayes 0.0.3
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/.document +5 -0
- data/.gitignore +21 -0
- data/LICENSE +20 -0
- data/README.rdoc +53 -0
- data/Rakefile +45 -0
- data/VERSION +1 -0
- data/lib/naive_bayes.rb +98 -0
- data/naive_bayes.gemspec +56 -0
- data/spec/db/naive.nb +1 -0
- data/spec/naive_bayes_spec.rb +75 -0
- data/spec/spec.opts +1 -0
- data/spec/spec_helper.rb +9 -0
- metadata +73 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA256:
|
3
|
+
metadata.gz: e76c76d77d5b450025bae114082124feaeb0adc06da2ff80135f0d20ec7911ca
|
4
|
+
data.tar.gz: 7b7b16848ed4cea466afcfaae2eb644c2417c0b052958c21858d2b71b18d49ea
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: a26f0e3250f6724091c117c6c43f33a681a8c611a52971d131b86fed3cdf4a55245604de5d4850ea18c70858b3ba60be5cdb28ed18c2cabe5f2de06604c6a9a7
|
7
|
+
data.tar.gz: 2e590688e26cd3481ddb5a48f3d08bc36d6fbd1fef2536c67b774386d287c6820f4eca14781e9c12147fd56b807b8909962ddc65868ce839dc3d40c2257fcf5b
|
data/.document
ADDED
data/.gitignore
ADDED
data/LICENSE
ADDED
@@ -0,0 +1,20 @@
|
|
1
|
+
Copyright (c) 2009 reddavis
|
2
|
+
|
3
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
4
|
+
a copy of this software and associated documentation files (the
|
5
|
+
"Software"), to deal in the Software without restriction, including
|
6
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
7
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
8
|
+
permit persons to whom the Software is furnished to do so, subject to
|
9
|
+
the following conditions:
|
10
|
+
|
11
|
+
The above copyright notice and this permission notice shall be
|
12
|
+
included in all copies or substantial portions of the Software.
|
13
|
+
|
14
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
15
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
16
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
17
|
+
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
18
|
+
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
19
|
+
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
20
|
+
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/README.rdoc
ADDED
@@ -0,0 +1,53 @@
|
|
1
|
+
= Naive Bayes Classifier
|
2
|
+
|
3
|
+
This is an extremely simple, straight forward Naive Bayes implementation.
|
4
|
+
|
5
|
+
== Install
|
6
|
+
|
7
|
+
gem sources -a -http://gemcutter.org
|
8
|
+
sudo gem install naive_bayes
|
9
|
+
|
10
|
+
== How To Use
|
11
|
+
|
12
|
+
require 'rubygems'
|
13
|
+
require 'naive_bayes'
|
14
|
+
|
15
|
+
a = NaiveBayes.new(:spam, :ham)
|
16
|
+
|
17
|
+
a.train(:spam, 'bad', 'word')
|
18
|
+
a.train(:ham, 'good', 'word')
|
19
|
+
|
20
|
+
b = "this is a bad sentence".split(' ')
|
21
|
+
|
22
|
+
a.classify(*b)
|
23
|
+
#=> [:spam, 0.03125]
|
24
|
+
|
25
|
+
You can also tell your classifier to save itself, so its easy for you to pick up where you left off:
|
26
|
+
|
27
|
+
require 'rubygems'
|
28
|
+
require 'naive_bayes'
|
29
|
+
|
30
|
+
a = NaiveBayes.new(:spam, :ham)
|
31
|
+
a.db_filepath = 'path/to/anywhere.nb'
|
32
|
+
|
33
|
+
a.train(:spam, 'bad', 'word')
|
34
|
+
a.train(:ham, 'good', 'word')
|
35
|
+
|
36
|
+
a.save
|
37
|
+
|
38
|
+
Some time goes past and we want to classify a new document we just received...
|
39
|
+
|
40
|
+
require 'rubygems'
|
41
|
+
require 'naive_bayes'
|
42
|
+
|
43
|
+
a = NaiveBayes.load('path/to/file')
|
44
|
+
|
45
|
+
b = "this is a bad sentence".split(' ')
|
46
|
+
|
47
|
+
# It's as if we were never apart
|
48
|
+
a.classify(*b)
|
49
|
+
#=> [:spam, 0.03125]
|
50
|
+
|
51
|
+
== Copyright
|
52
|
+
|
53
|
+
Copyright (c) 2009 Red Davis. See LICENSE for details.
|
data/Rakefile
ADDED
@@ -0,0 +1,45 @@
|
|
1
|
+
require 'rubygems'
|
2
|
+
require 'rake'
|
3
|
+
|
4
|
+
begin
|
5
|
+
require 'jeweler'
|
6
|
+
Jeweler::Tasks.new do |gem|
|
7
|
+
gem.name = "naive_bayes"
|
8
|
+
gem.summary = %Q{Simple straight forward Naive Bayes classifier implementation}
|
9
|
+
gem.description = %Q{Simple straight forward Naive Bayes classifier implementation}
|
10
|
+
gem.email = "reddavis@gmail.com"
|
11
|
+
gem.homepage = "http://github.com/reddavis/Naive-Bayes"
|
12
|
+
gem.authors = ["reddavis"]
|
13
|
+
gem.add_development_dependency "rspec", ">= 1.2.9"
|
14
|
+
# gem is a Gem::Specification... see http://www.rubygems.org/read/chapter/20 for additional settings
|
15
|
+
end
|
16
|
+
Jeweler::GemcutterTasks.new
|
17
|
+
rescue LoadError
|
18
|
+
puts "Jeweler (or a dependency) not available. Install it with: sudo gem install jeweler"
|
19
|
+
end
|
20
|
+
|
21
|
+
require 'spec/rake/spectask'
|
22
|
+
Spec::Rake::SpecTask.new(:spec) do |spec|
|
23
|
+
spec.libs << 'lib' << 'spec'
|
24
|
+
spec.spec_files = FileList['spec/**/*_spec.rb']
|
25
|
+
end
|
26
|
+
|
27
|
+
Spec::Rake::SpecTask.new(:rcov) do |spec|
|
28
|
+
spec.libs << 'lib' << 'spec'
|
29
|
+
spec.pattern = 'spec/**/*_spec.rb'
|
30
|
+
spec.rcov = true
|
31
|
+
end
|
32
|
+
|
33
|
+
task :spec => :check_dependencies
|
34
|
+
|
35
|
+
task :default => :spec
|
36
|
+
|
37
|
+
require 'rake/rdoctask'
|
38
|
+
Rake::RDocTask.new do |rdoc|
|
39
|
+
version = File.exist?('VERSION') ? File.read('VERSION') : ""
|
40
|
+
|
41
|
+
rdoc.rdoc_dir = 'rdoc'
|
42
|
+
rdoc.title = "naive_bayes #{version}"
|
43
|
+
rdoc.rdoc_files.include('README*')
|
44
|
+
rdoc.rdoc_files.include('lib/**/*.rb')
|
45
|
+
end
|
data/VERSION
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
0.0.3
|
data/lib/naive_bayes.rb
ADDED
@@ -0,0 +1,98 @@
|
|
1
|
+
# Bayes Theorem
|
2
|
+
# P(A|B) = P(B|A) * P(A) / P(B)
|
3
|
+
|
4
|
+
# Terminology
|
5
|
+
# An ITEM is made up of FEATURES
|
6
|
+
# An ITEM belongs to a CLASS
|
7
|
+
|
8
|
+
# Bayes With Our Terminology
|
9
|
+
# P(Class | Item) = P(Item | Class) * P(Class) / P(Item)
|
10
|
+
|
11
|
+
# However, when classifying, P(Item) is the same across all calcualtions
|
12
|
+
# So we don't bother to calculate it
|
13
|
+
class NaiveBayes
|
14
|
+
|
15
|
+
class << self
|
16
|
+
def load(db_path)
|
17
|
+
data = ""
|
18
|
+
File.open(db_path) do |f|
|
19
|
+
while line = f.gets
|
20
|
+
data << line
|
21
|
+
end
|
22
|
+
end
|
23
|
+
Marshal.load(data)
|
24
|
+
end
|
25
|
+
end
|
26
|
+
|
27
|
+
attr_accessor :db_filepath
|
28
|
+
|
29
|
+
def initialize(*klasses)
|
30
|
+
@features_count = {}
|
31
|
+
@klass_count = {}
|
32
|
+
@klasses = klasses
|
33
|
+
|
34
|
+
klasses.each do |klass|
|
35
|
+
@features_count[klass] = Hash.new(0.0)
|
36
|
+
@klass_count[klass] = 0.0
|
37
|
+
end
|
38
|
+
end
|
39
|
+
|
40
|
+
def train(klass, *features)
|
41
|
+
features.uniq.each do |feature|
|
42
|
+
@features_count[klass][feature] += 1
|
43
|
+
end
|
44
|
+
@klass_count[klass] += 1
|
45
|
+
end
|
46
|
+
|
47
|
+
def untrain(klass, *features)
|
48
|
+
features.uniq.each do |feature|
|
49
|
+
@features_count[klass][feature] -= 1
|
50
|
+
end
|
51
|
+
@klass_count[klass] -= 1
|
52
|
+
end
|
53
|
+
|
54
|
+
#P(Class | Item) = P(Item | Class) * P(Class)
|
55
|
+
def classify(*features)
|
56
|
+
scores = {}
|
57
|
+
@klasses.each do |klass|
|
58
|
+
scores[klass] = (prob_of_item_given_a_class(features, klass) * prob_of_class(klass))
|
59
|
+
end
|
60
|
+
return [] if scores.values.reduce(:+) == 0.0
|
61
|
+
scores.sort {|a,b| b[1] <=> a[1]}[0]
|
62
|
+
end
|
63
|
+
|
64
|
+
def save
|
65
|
+
raise "You haven't set a db_filpath, I dont know where to save" if @db_filepath.nil?
|
66
|
+
File.open(@db_filepath, "w+") do |f|
|
67
|
+
f.write(Marshal.dump(self))
|
68
|
+
end
|
69
|
+
end
|
70
|
+
|
71
|
+
private
|
72
|
+
|
73
|
+
# P(Item | Class)
|
74
|
+
def prob_of_item_given_a_class(features, klass)
|
75
|
+
a = features.inject(1.0) do |sum, feature|
|
76
|
+
prob = prob_of_feature_given_a_class(feature, klass)
|
77
|
+
end
|
78
|
+
end
|
79
|
+
|
80
|
+
# P(Feature | Class)
|
81
|
+
def prob_of_feature_given_a_class(feature, klass)
|
82
|
+
# If there is not any sentence in our trained models we return nil value
|
83
|
+
@feature = @features_count[klass][feature.to_sym]
|
84
|
+
feature.nil? ? feature : @features_count[klass][feature.to_sym] / @klass_count[klass]
|
85
|
+
end
|
86
|
+
|
87
|
+
# P(Class)
|
88
|
+
def prob_of_class(klass)
|
89
|
+
@klass_count[klass] / total_items
|
90
|
+
end
|
91
|
+
|
92
|
+
def total_items
|
93
|
+
@klass_count.inject(0) do |sum, klass|
|
94
|
+
sum += klass[1]
|
95
|
+
end
|
96
|
+
end
|
97
|
+
|
98
|
+
end
|
data/naive_bayes.gemspec
ADDED
@@ -0,0 +1,56 @@
|
|
1
|
+
# Generated by jeweler
|
2
|
+
# DO NOT EDIT THIS FILE DIRECTLY
|
3
|
+
# Instead, edit Jeweler::Tasks in Rakefile, and run the gemspec command
|
4
|
+
# -*- encoding: utf-8 -*-
|
5
|
+
|
6
|
+
Gem::Specification.new do |s|
|
7
|
+
s.name = %q{solid_naive_bayes}
|
8
|
+
s.version = "0.0.3"
|
9
|
+
|
10
|
+
s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
|
11
|
+
s.authors = ["reddavis"]
|
12
|
+
s.date = %q{2010-03-11}
|
13
|
+
s.description = %q{Simple straight forward Naive Bayes classifier implementation}
|
14
|
+
s.email = %q{reddavis@gmail.com}
|
15
|
+
s.extra_rdoc_files = [
|
16
|
+
"LICENSE",
|
17
|
+
"README.rdoc"
|
18
|
+
]
|
19
|
+
s.files = [
|
20
|
+
".document",
|
21
|
+
".gitignore",
|
22
|
+
"LICENSE",
|
23
|
+
"README.rdoc",
|
24
|
+
"Rakefile",
|
25
|
+
"VERSION",
|
26
|
+
"lib/naive_bayes.rb",
|
27
|
+
"naive_bayes.gemspec",
|
28
|
+
"spec/db/naive.nb",
|
29
|
+
"spec/naive_bayes_spec.rb",
|
30
|
+
"spec/spec.opts",
|
31
|
+
"spec/spec_helper.rb"
|
32
|
+
]
|
33
|
+
s.homepage = %q{http://github.com/reddavis/Naive-Bayes}
|
34
|
+
s.rdoc_options = ["--charset=UTF-8"]
|
35
|
+
s.require_paths = ["lib"]
|
36
|
+
s.rubygems_version = %q{1.3.6}
|
37
|
+
s.summary = %q{Simple straight forward Naive Bayes classifier implementation}
|
38
|
+
s.test_files = [
|
39
|
+
"spec/naive_bayes_spec.rb",
|
40
|
+
"spec/spec_helper.rb"
|
41
|
+
]
|
42
|
+
|
43
|
+
if s.respond_to? :specification_version then
|
44
|
+
current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
|
45
|
+
s.specification_version = 3
|
46
|
+
|
47
|
+
if Gem::Version.new(Gem::RubyGemsVersion) >= Gem::Version.new('1.2.0') then
|
48
|
+
s.add_development_dependency(%q<rspec>, [">= 1.2.9"])
|
49
|
+
else
|
50
|
+
s.add_dependency(%q<rspec>, [">= 1.2.9"])
|
51
|
+
end
|
52
|
+
else
|
53
|
+
s.add_dependency(%q<rspec>, [">= 1.2.9"])
|
54
|
+
end
|
55
|
+
end
|
56
|
+
|
data/spec/db/naive.nb
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
o:NaiveBayes :@features_count{: spam}I"bad:EFf1I" word;Ff1f0:ham}I"we;Ff1I"bad;Ff1@:@klass_count{;f1; f1:
|
@@ -0,0 +1,75 @@
|
|
1
|
+
require File.expand_path(File.dirname(__FILE__) + '/spec_helper')
|
2
|
+
|
3
|
+
describe "NaiveBayes" do
|
4
|
+
|
5
|
+
describe "Classification" do
|
6
|
+
before do
|
7
|
+
@classifier = create_and_train_classifier
|
8
|
+
end
|
9
|
+
|
10
|
+
it "should classify as spam with a score of 0.5" do
|
11
|
+
a = @classifier.classify('bad', 'word')
|
12
|
+
a[0].should == :spam
|
13
|
+
a[1].should == 0.5
|
14
|
+
end
|
15
|
+
|
16
|
+
it "should change information" do
|
17
|
+
@classifier.untrain(:spam, 'bad', 'word')
|
18
|
+
a = @classifier.classify('bad', 'word')
|
19
|
+
a[0].should == :ham
|
20
|
+
a[1].should == 1.0
|
21
|
+
end
|
22
|
+
end
|
23
|
+
|
24
|
+
describe "Saving the NB" do
|
25
|
+
describe "DB filepath has been set" do
|
26
|
+
before do
|
27
|
+
@classifier = NaiveBayes.new(:spam, :ham)
|
28
|
+
@classifier.db_filepath = db_filepath
|
29
|
+
end
|
30
|
+
|
31
|
+
it "should save to the filepath provided" do
|
32
|
+
FileUtils.rm(db_filepath, :force => true)
|
33
|
+
@classifier.save
|
34
|
+
File.exists?(db_filepath).should be_true
|
35
|
+
end
|
36
|
+
end
|
37
|
+
|
38
|
+
describe "DB filepath has no been set" do
|
39
|
+
it "should raise an error" do
|
40
|
+
lambda do
|
41
|
+
NaiveBayes.new(:spam, :ham).save
|
42
|
+
end.should raise_error
|
43
|
+
end
|
44
|
+
end
|
45
|
+
end
|
46
|
+
|
47
|
+
describe "Load" do
|
48
|
+
before do
|
49
|
+
classifier = NaiveBayes.new(:spam, :ham)
|
50
|
+
classifier.db_filepath = db_filepath
|
51
|
+
classifier.train(:spam, 'bad', 'word')
|
52
|
+
classifier.train(:ham, 'we', 'bad')
|
53
|
+
classifier.save
|
54
|
+
end
|
55
|
+
|
56
|
+
it "should return 0.5" do
|
57
|
+
classifier = NaiveBayes.load(db_filepath)
|
58
|
+
classifier.classify('bad', 'word')[1].should == 0.5
|
59
|
+
end
|
60
|
+
end
|
61
|
+
|
62
|
+
private
|
63
|
+
|
64
|
+
def create_and_train_classifier
|
65
|
+
a = NaiveBayes.new(:spam, :ham)
|
66
|
+
a.train(:spam, 'bad', 'word')
|
67
|
+
a.train(:ham, 'we', 'bad')
|
68
|
+
a
|
69
|
+
end
|
70
|
+
|
71
|
+
def db_filepath
|
72
|
+
File.expand_path(File.dirname(__FILE__) + '/db/naive.nb')
|
73
|
+
end
|
74
|
+
|
75
|
+
end
|
data/spec/spec.opts
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
--color
|
data/spec/spec_helper.rb
ADDED
metadata
ADDED
@@ -0,0 +1,73 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: solid_naive_bayes
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.3
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- reddavis
|
8
|
+
autorequire:
|
9
|
+
bindir: bin
|
10
|
+
cert_chain: []
|
11
|
+
date: 2010-03-11 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
name: rspec
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - ">="
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: 1.2.9
|
20
|
+
type: :development
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - ">="
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: 1.2.9
|
27
|
+
description: Simple straight forward Naive Bayes classifier implementation
|
28
|
+
email: reddavis@gmail.com
|
29
|
+
executables: []
|
30
|
+
extensions: []
|
31
|
+
extra_rdoc_files:
|
32
|
+
- LICENSE
|
33
|
+
- README.rdoc
|
34
|
+
files:
|
35
|
+
- ".document"
|
36
|
+
- ".gitignore"
|
37
|
+
- LICENSE
|
38
|
+
- README.rdoc
|
39
|
+
- Rakefile
|
40
|
+
- VERSION
|
41
|
+
- lib/naive_bayes.rb
|
42
|
+
- naive_bayes.gemspec
|
43
|
+
- spec/db/naive.nb
|
44
|
+
- spec/naive_bayes_spec.rb
|
45
|
+
- spec/spec.opts
|
46
|
+
- spec/spec_helper.rb
|
47
|
+
homepage: http://github.com/reddavis/Naive-Bayes
|
48
|
+
licenses: []
|
49
|
+
metadata: {}
|
50
|
+
post_install_message:
|
51
|
+
rdoc_options:
|
52
|
+
- "--charset=UTF-8"
|
53
|
+
require_paths:
|
54
|
+
- lib
|
55
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
56
|
+
requirements:
|
57
|
+
- - ">="
|
58
|
+
- !ruby/object:Gem::Version
|
59
|
+
version: '0'
|
60
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
61
|
+
requirements:
|
62
|
+
- - ">="
|
63
|
+
- !ruby/object:Gem::Version
|
64
|
+
version: '0'
|
65
|
+
requirements: []
|
66
|
+
rubyforge_project:
|
67
|
+
rubygems_version: 3.0.0.beta3
|
68
|
+
signing_key:
|
69
|
+
specification_version: 3
|
70
|
+
summary: Simple straight forward Naive Bayes classifier implementation
|
71
|
+
test_files:
|
72
|
+
- spec/naive_bayes_spec.rb
|
73
|
+
- spec/spec_helper.rb
|