bayes_naive_jdp 0.0.1 → 0.0.2
Sign up to get free protection for your applications and to get access to all the features.
- data/README.md +30 -9
- data/lib/bayes_naive_jdp/version.rb +1 -1
- data/test/test_misc.rb +40 -0
- metadata +5 -3
data/README.md
CHANGED
@@ -1,6 +1,14 @@
|
|
1
1
|
# BayesNaiveJdp
|
2
2
|
|
3
|
-
|
3
|
+
This is a very simple Naive Bayesian classifier, build just for fun.
|
4
|
+
|
5
|
+
I decided to use it as an example while learning to package ruby code.
|
6
|
+
|
7
|
+
The algorithm used here is not original, but an adaptation from Burak Kanber's
|
8
|
+
[Machine Learning in Javascript](http://readable.cc/feed/view/34236/burak-kanber-s-blog) series.
|
9
|
+
|
10
|
+
|
11
|
+
|
4
12
|
|
5
13
|
## Installation
|
6
14
|
|
@@ -18,12 +26,25 @@ Or install it yourself as:
|
|
18
26
|
|
19
27
|
## Usage
|
20
28
|
|
21
|
-
|
22
|
-
|
23
|
-
|
29
|
+
# create an instance of the Classifier
|
30
|
+
classifier = BayesNaiveJdp::Classifier.new
|
31
|
+
|
32
|
+
# Train the Classifier using many samples from each valid category
|
33
|
+
# documents are strings, categories are strings, symbols, or integers
|
34
|
+
classifier.train(document_a_1, category_a)
|
35
|
+
classifier.train(document_a_2, category_a)
|
36
|
+
classifier.train(document_a_3, category_a)
|
37
|
+
..
|
38
|
+
classifier.train(document_b_1, category_b)
|
39
|
+
classifier.train(document_b_2, category_b)
|
40
|
+
classifier.train(document_b_3, category_b)
|
41
|
+
..
|
42
|
+
|
43
|
+
# use the classifier to categorize an unknown document
|
44
|
+
results = classifier.classify(mystery_document)
|
45
|
+
category = results[:winner][:classification]
|
46
|
+
confidence = results[:winner][:confidence]
|
47
|
+
|
48
|
+
# also, results[:all_scores] is a hash
|
49
|
+
# with categories as indices and confidence levels as values
|
24
50
|
|
25
|
-
1. Fork it
|
26
|
-
2. Create your feature branch (`git checkout -b my-new-feature`)
|
27
|
-
3. Commit your changes (`git commit -am 'Add some feature'`)
|
28
|
-
4. Push to the branch (`git push origin my-new-feature`)
|
29
|
-
5. Create new Pull Request
|
data/test/test_misc.rb
ADDED
@@ -0,0 +1,40 @@
|
|
1
|
+
require 'pathname'
|
2
|
+
require 'test/unit'
|
3
|
+
|
4
|
+
dir = Pathname.new File.expand_path(File.dirname(__FILE__))
|
5
|
+
require dir + '..' + 'lib' + 'bayes_naive_jdp'
|
6
|
+
|
7
|
+
DOCUMENT_PATH = dir + 'documents'
|
8
|
+
CONFIDENCE_MIN = 0.98
|
9
|
+
|
10
|
+
class MiscTest < Test::Unit::TestCase
|
11
|
+
def test_integer_categories
|
12
|
+
categories = [1, 2, 3]
|
13
|
+
classifier = BayesNaiveJdp::Classifier.new
|
14
|
+
categories.each do |cat|
|
15
|
+
4.times do |i|
|
16
|
+
classifier.train("#{cat} " * i, cat)
|
17
|
+
end
|
18
|
+
end
|
19
|
+
|
20
|
+
categories.each do |cat|
|
21
|
+
answer = classifier.classify("#{cat} " * 5)
|
22
|
+
assert_equal(answer[:winner][:classification], cat)
|
23
|
+
end
|
24
|
+
end
|
25
|
+
|
26
|
+
def test_symbol_categories_
|
27
|
+
categories = [:one, :two, :three]
|
28
|
+
classifier = BayesNaiveJdp::Classifier.new
|
29
|
+
categories.each do |cat|
|
30
|
+
4.times do |i|
|
31
|
+
classifier.train("#{cat.to_s} " * i, cat)
|
32
|
+
end
|
33
|
+
end
|
34
|
+
|
35
|
+
categories.each do |cat|
|
36
|
+
answer = classifier.classify("#{cat.to_s} " * 5)
|
37
|
+
assert_equal(answer[:winner][:classification], cat)
|
38
|
+
end
|
39
|
+
end
|
40
|
+
end
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: bayes_naive_jdp
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.0.
|
4
|
+
version: 0.0.2
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -77,6 +77,7 @@ files:
|
|
77
77
|
- test/documents/schneier-4
|
78
78
|
- test/documents/schneier-5
|
79
79
|
- test/test_author_classification.rb
|
80
|
+
- test/test_misc.rb
|
80
81
|
homepage: ''
|
81
82
|
licenses:
|
82
83
|
- MIT
|
@@ -92,7 +93,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
|
|
92
93
|
version: '0'
|
93
94
|
segments:
|
94
95
|
- 0
|
95
|
-
hash:
|
96
|
+
hash: -4046118477202807926
|
96
97
|
required_rubygems_version: !ruby/object:Gem::Requirement
|
97
98
|
none: false
|
98
99
|
requirements:
|
@@ -101,7 +102,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
101
102
|
version: '0'
|
102
103
|
segments:
|
103
104
|
- 0
|
104
|
-
hash:
|
105
|
+
hash: -4046118477202807926
|
105
106
|
requirements: []
|
106
107
|
rubyforge_project:
|
107
108
|
rubygems_version: 1.8.25
|
@@ -125,3 +126,4 @@ test_files:
|
|
125
126
|
- test/documents/schneier-4
|
126
127
|
- test/documents/schneier-5
|
127
128
|
- test/test_author_classification.rb
|
129
|
+
- test/test_misc.rb
|