jnb_classifier 0.4.1 → 0.4.2
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/Gemfile.lock +1 -1
- data/README.md +65 -4
- data/README.md.backup +80 -0
- data/lib/jnb_classifier.rb +15 -16
- data/lib/jnb_classifier/version.rb +1 -1
- metadata +3 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 99aaba58f7a9e3c702d91490ced2e45f1413e252
|
4
|
+
data.tar.gz: 7430e3ab70b3aa959c43e5a518e4323878b65378
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: b1365b424c12f58ede00c7c44060808daf52e2b2ea41f7ed74fbe7ea5aacb1b735c9d0126ecea9220b7618768d370407d6d72cc5195ac7924363486b33ff8d58
|
7
|
+
data.tar.gz: 34781f0bef1ab8d812eda44dbb3be5b33c1e37110dfe4a8dbbfa560ddbb7cbfd5781067c1eb6219c1e88c13206def3381cbd5ce8eb00d9604145a412ede6cc17
|
data/Gemfile.lock
CHANGED
data/README.md
CHANGED
@@ -1,12 +1,40 @@
|
|
1
1
|
# JnbClassifier
|
2
2
|
|
3
|
-
|
3
|
+
Naive Bayes classifier for japanse text.
|
4
|
+
|
5
|
+
## Description
|
6
|
+
JnbClassifier helps you classify japanese text.
|
7
|
+
|
8
|
+
You need Morphological Analysis Tool, Mecab and the dictionary to use JnbClassifier.
|
9
|
+
|
10
|
+
If you use linux(ubuntu), you can install Mecab and the dictionary as follows.
|
11
|
+
|
12
|
+
1:Mecab
|
13
|
+
download site: https://drive.google.com/uc?export=download&id=0B4y35FiV1wh7cENtOXlicTFaRUE
|
14
|
+
|
15
|
+
```bash
|
16
|
+
$ cd mecab-X.X
|
17
|
+
$ ./configure
|
18
|
+
$ make
|
19
|
+
$ make check
|
20
|
+
$ sudo make install
|
21
|
+
```
|
22
|
+
|
23
|
+
2:the dictionary for Mecab
|
24
|
+
download site: https://drive.google.com/uc?export=download&id=0B4y35FiV1wh7MWVlSDBCSXZMTXM
|
25
|
+
|
26
|
+
```bash
|
27
|
+
$ tar zxfv mecab-ipadic-2.7.0-XXXX.tar.gz
|
28
|
+
$ cd mecab-ipadic-2.7.0-XXXX
|
29
|
+
$ ./configure --with-charset=utf8
|
30
|
+
$ make
|
31
|
+
$ sudo make install
|
32
|
+
```
|
4
33
|
|
5
|
-
TODO: Delete this and the text above, and describe your gem
|
6
34
|
|
7
35
|
## Installation
|
8
36
|
|
9
|
-
Add this line to your application's Gemfile
|
37
|
+
Add this line to your application's Gemfile
|
10
38
|
|
11
39
|
```ruby
|
12
40
|
gem 'jnb_classifier'
|
@@ -22,7 +50,40 @@ Or install it yourself as:
|
|
22
50
|
|
23
51
|
## Usage
|
24
52
|
|
25
|
-
|
53
|
+
#### 1: Generate Document(label, attributes) from text
|
54
|
+
First you need create documents which neeed two parameters: category name and text.
|
55
|
+
|
56
|
+
```
|
57
|
+
document1 = JnbClassifier::Document.new('hayao','私の住んでいる場所は風の谷ではありません。')
|
58
|
+
#=> @label="hayao", @attributes={"私"=>1, "場所"=>1, "風"=>1, "谷"=>1}
|
59
|
+
document2 = JnbClassifier::Document.new('hayao','私は魔女の宅急便が大好きです')
|
60
|
+
#=> @label="hayao", @attributes={"私"=>1, "魔女"=>1, "宅急便"=>1, "大好き"=>1}
|
61
|
+
document3 = JnbClassifier::Document.new('hujiko','僕、ドラえもん。')
|
62
|
+
#=> @label="hujiko", @attributes={"僕"=>1, "ドラえもん"=>1}
|
63
|
+
```
|
64
|
+
|
65
|
+
#### 2: Learn documents
|
66
|
+
Your classifier can learn from documents you created.
|
67
|
+
|
68
|
+
And the more you give apropriate samples, the smarter your classifier will be.
|
69
|
+
|
70
|
+
```
|
71
|
+
nb = nb = JnbClassifier::Classifier.new
|
72
|
+
mb.learn(document1)
|
73
|
+
mb.learn(document2)
|
74
|
+
mb.learn(document3)
|
75
|
+
```
|
76
|
+
|
77
|
+
#### 3:Classify
|
78
|
+
When you input attributes(hash), ,classify method return the category which is selected by Naive Bayes.
|
79
|
+
|
80
|
+
```
|
81
|
+
nb.classify({"私"=>1, "魔女"=>1, "宅急便"=>1} )
|
82
|
+
#=> "hayao"
|
83
|
+
nb.result
|
84
|
+
#=> {"hayao"=>-16.72902719126572, "hujiko"=>-20.435583764494627} # the values is logarithm.
|
85
|
+
```
|
86
|
+
|
26
87
|
|
27
88
|
## Development
|
28
89
|
|
data/README.md.backup
ADDED
@@ -0,0 +1,80 @@
|
|
1
|
+
# JnbClassifier
|
2
|
+
|
3
|
+
Naive Bayes classifier for japanse text.
|
4
|
+
|
5
|
+
## Description
|
6
|
+
JnbClassifier help you classify japanese text.
|
7
|
+
|
8
|
+
You need Morphological Analysis Tool, Mecab and the dictionary to use JnbClassifier.
|
9
|
+
|
10
|
+
If you use linux(ubuntu), you can install Mecab and the dictionary as follows.
|
11
|
+
|
12
|
+
1:Mecab
|
13
|
+
download site: https://drive.google.com/uc?export=download&id=0B4y35FiV1wh7cENtOXlicTFaRUE
|
14
|
+
|
15
|
+
```bash
|
16
|
+
$ cd mecab-X.X
|
17
|
+
$ ./configure
|
18
|
+
$ make
|
19
|
+
$ make check
|
20
|
+
$ sudo make install
|
21
|
+
```
|
22
|
+
|
23
|
+
2:the dictionary for Mecab
|
24
|
+
download site: https://drive.google.com/uc?export=download&id=0B4y35FiV1wh7MWVlSDBCSXZMTXM
|
25
|
+
|
26
|
+
```bash
|
27
|
+
$ tar zxfv mecab-ipadic-2.7.0-XXXX.tar.gz
|
28
|
+
$ cd mecab-ipadic-2.7.0-XXXX
|
29
|
+
$ ./configure --with-charset=utf8
|
30
|
+
$ make
|
31
|
+
$ sudo make install
|
32
|
+
```
|
33
|
+
|
34
|
+
|
35
|
+
## Installation
|
36
|
+
|
37
|
+
Add this line to your application's Gemfile
|
38
|
+
|
39
|
+
```ruby
|
40
|
+
gem 'jnb_classifier'
|
41
|
+
```
|
42
|
+
|
43
|
+
And then execute:
|
44
|
+
|
45
|
+
$ bundle
|
46
|
+
|
47
|
+
Or install it yourself as:
|
48
|
+
|
49
|
+
$ gem install jnb_classifier
|
50
|
+
|
51
|
+
## Usage
|
52
|
+
|
53
|
+
#### 1: generate hash from text
|
54
|
+
|
55
|
+
```
|
56
|
+
document1 = JnbClassifier::Document.new('hayao','')
|
57
|
+
|
58
|
+
```
|
59
|
+
|
60
|
+
|
61
|
+
|
62
|
+
|
63
|
+
|
64
|
+
## Development
|
65
|
+
|
66
|
+
After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake spec` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.
|
67
|
+
|
68
|
+
To install this gem onto your local machine, run `bundle exec rake install`. To release a new version, update the version number in `version.rb`, and then run `bundle exec rake release`, which will create a git tag for the version, push git commits and tags, and push the `.gem` file to [rubygems.org](https://rubygems.org).
|
69
|
+
|
70
|
+
## Contributing
|
71
|
+
|
72
|
+
Bug reports and pull requests are welcome on GitHub at https://github.com/[USERNAME]/jnb_classifier. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the [Contributor Covenant](http://contributor-covenant.org) code of conduct.
|
73
|
+
|
74
|
+
## License
|
75
|
+
|
76
|
+
The gem is available as open source under the terms of the [MIT License](https://opensource.org/licenses/MIT).
|
77
|
+
|
78
|
+
## Code of Conduct
|
79
|
+
|
80
|
+
Everyone interacting in the JnbClassifier project’s codebases, issue trackers, chat rooms and mailing lists is expected to follow the [code of conduct](https://github.com/[USERNAME]/jnb_classifier/blob/master/CODE_OF_CONDUCT.md).
|
data/lib/jnb_classifier.rb
CHANGED
@@ -7,8 +7,8 @@ module JnbClassifier
|
|
7
7
|
attr_reader :result
|
8
8
|
|
9
9
|
def initialize
|
10
|
-
@frequency_table = Hash.new
|
11
|
-
@word_table = Hash.new
|
10
|
+
@frequency_table = Hash.new # frequency table for each class
|
11
|
+
@word_table = Hash.new # word feature table
|
12
12
|
@label_count = Hash.new(0) # count by each label
|
13
13
|
@total_count = 0 # total learned documents
|
14
14
|
@result = Hash.new
|
@@ -36,13 +36,14 @@ module JnbClassifier
|
|
36
36
|
|
37
37
|
# P(Label)
|
38
38
|
@label_count.each{|label,freq|
|
39
|
-
label_p[label] = Math.log(freq.
|
39
|
+
label_p[label] = Math.log(freq.fdiv(@total_count))
|
40
40
|
}
|
41
41
|
|
42
|
-
# P(X|Label)
|
42
|
+
# P(X|Label)
|
43
43
|
@frequency_table.each_key{|label|
|
44
|
+
deno = @label_count[label] + @word_table.size()
|
44
45
|
@word_table.each_key{|word|
|
45
|
-
laplace_word_p[label] += Math.log((@frequency_table[label][word] + 1).
|
46
|
+
laplace_word_p[label] += Math.log( (@frequency_table[label][word] + 1).fdiv(deno) )
|
46
47
|
}
|
47
48
|
score[label] = laplace_word_p[label] + label_p[label]
|
48
49
|
}
|
@@ -51,7 +52,7 @@ module JnbClassifier
|
|
51
52
|
score.each{|label, value|
|
52
53
|
@result[label] = value
|
53
54
|
}
|
54
|
-
score.
|
55
|
+
score.max_by{ |x| x[1] }
|
55
56
|
end
|
56
57
|
end
|
57
58
|
|
@@ -60,23 +61,21 @@ module JnbClassifier
|
|
60
61
|
attr_reader :label
|
61
62
|
attr_reader :attributes
|
62
63
|
|
63
|
-
def initialize(label,
|
64
|
+
def initialize(label,doc)
|
64
65
|
@label = label # String
|
65
|
-
@attributes = create_attributes(
|
66
|
+
@attributes = create_attributes(doc) # Hsah
|
66
67
|
end
|
67
68
|
|
68
|
-
def create_attributes(
|
69
|
+
def create_attributes(doc)
|
69
70
|
attributes = Hash.new(0)
|
70
|
-
|
71
|
-
|
72
|
-
|
73
|
-
|
74
|
-
attributes[n.surface] += 1 if n.feature.match(/名詞/)
|
75
|
-
end
|
76
|
-
}
|
71
|
+
nm = Natto::MeCab.new
|
72
|
+
nm.parse(doc) do |n|
|
73
|
+
attributes[n.surface] += 1 if n.feature.match(/名詞/)
|
74
|
+
end
|
77
75
|
attributes
|
78
76
|
end
|
79
77
|
end
|
80
78
|
|
81
79
|
end
|
82
80
|
|
81
|
+
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: jnb_classifier
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.4.
|
4
|
+
version: 0.4.2
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- chamao
|
8
8
|
autorequire:
|
9
9
|
bindir: exe
|
10
10
|
cert_chain: []
|
11
|
-
date: 2017-11-
|
11
|
+
date: 2017-11-05 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: bundler
|
@@ -81,6 +81,7 @@ files:
|
|
81
81
|
- Gemfile.lock
|
82
82
|
- LICENSE.txt
|
83
83
|
- README.md
|
84
|
+
- README.md.backup
|
84
85
|
- Rakefile
|
85
86
|
- bin/console
|
86
87
|
- bin/setup
|