known_item_search_classifier 0.1.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/lib/known_item_search_classifier/classifier.rb +53 -0
- data/lib/known_item_search_classifier/default_training_set.rb +21 -0
- data/lib/known_item_search_classifier/feature_extractor.rb +68 -0
- data/lib/known_item_search_classifier.rb +3 -0
- data/test/known_item_search_classifier_test.rb +62 -0
- metadata +106 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA1:
|
3
|
+
metadata.gz: df03a8dc0661439a41c7d3366a49573f463a176f
|
4
|
+
data.tar.gz: 58713a770bba4a173adc4e88bb507f256b4eafc5
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 470565b3b2932df41a0d02b99048746b5f5e97de476b273a9648d473d63d7e73e44d038ff2b35ffd1f0bbd413960d35df5ca1ad8e8b8902038aedb57d64225f8
|
7
|
+
data.tar.gz: 7e81f8262925a653ab33c12844947f9df40b5510cf3e43d801d6a31cab7f3bf3a3de58dc9693ecbd71283ff309dc9bdb6490d1a6be25fe436726ffc09fea28e0
|
@@ -0,0 +1,53 @@
|
|
1
|
+
# Classifies search strings as either known-item searches or unknown-item searches
|
2
|
+
require 'gaussian_naive_bayes'
|
3
|
+
|
4
|
+
module KnownItemSearchClassifier
|
5
|
+
class Classifier
|
6
|
+
def initialize
|
7
|
+
set = DefaultTrainingSet.new
|
8
|
+
@default_training_set = GaussianNaiveBayes::Classifier.new set.categories_summaries, set.categories_probabilities
|
9
|
+
end
|
10
|
+
def is_known_item_search? query_string
|
11
|
+
return classify query_string
|
12
|
+
end
|
13
|
+
def train training_set
|
14
|
+
if defined? @custom_training_set
|
15
|
+
@custom_training_set = GaussianNaiveBayes::Learner.new
|
16
|
+
end
|
17
|
+
training_set.each do |query|
|
18
|
+
submit_vector query
|
19
|
+
end
|
20
|
+
end
|
21
|
+
def train_from_csv filename
|
22
|
+
if defined? @custom_training_set
|
23
|
+
@custom_training_set = GaussianNaiveBayes::Learner.new
|
24
|
+
end
|
25
|
+
csv = CSV.read(filename)
|
26
|
+
csv.each do |line|
|
27
|
+
submit_vector line
|
28
|
+
end
|
29
|
+
end
|
30
|
+
|
31
|
+
private
|
32
|
+
def classify string
|
33
|
+
f = FeatureExtractor.new string
|
34
|
+
feature_array = f.feature_array
|
35
|
+
if defined? @custom_training_set
|
36
|
+
classifier = @custom_training_set.classifier
|
37
|
+
query_class = classifier.classify(feature_array)
|
38
|
+
else
|
39
|
+
query_class = @default_training_set.classify(feature_array)
|
40
|
+
end
|
41
|
+
return query_class
|
42
|
+
if :known == query_class
|
43
|
+
return true
|
44
|
+
else
|
45
|
+
return false
|
46
|
+
end
|
47
|
+
end
|
48
|
+
def submit_vector arr
|
49
|
+
f = FeatureExtractor.new arr[0]
|
50
|
+
@custom_training_set.train f.feature_array, arr[1]
|
51
|
+
end
|
52
|
+
end
|
53
|
+
end
|
@@ -0,0 +1,21 @@
|
|
1
|
+
module KnownItemSearchClassifier
|
2
|
+
class DefaultTrainingSet
|
3
|
+
attr_reader :categories_probabilities, :categories_summaries
|
4
|
+
def initialize
|
5
|
+
@categories_probabilities={:unknown=>0.78, :known=>0.22}
|
6
|
+
@categories_summaries= {
|
7
|
+
:unknown=>{
|
8
|
+
0=>{:mean=>0.2564102564102564, :standard_deviation=>0.4394771815921655},
|
9
|
+
1=>{:mean=>0.03418803418803419, :standard_deviation=>0.11344969312798027},
|
10
|
+
2=>{:mean=>0.002564102564102564, :standard_deviation=>0.0226455406828919},
|
11
|
+
3=>{:mean=>0.12991452991452992, :standard_deviation=>0.26648206508636013},
|
12
|
+
4=>{:mean=>2.7948717948717947, :standard_deviation=>2.053561836691609}},
|
13
|
+
:known=>{
|
14
|
+
0=>{:mean=>0.5454545454545454, :standard_deviation=>0.5096471914376255},
|
15
|
+
1=>{:mean=>0.051659451659451655, :standard_deviation=>0.07957404805575267},
|
16
|
+
2=>{:mean=>0.021248196248196245, :standard_deviation=>0.04412470821426937},
|
17
|
+
3=>{:mean=>0.22550505050505054, :standard_deviation=>0.2520704609787127},
|
18
|
+
4=>{:mean=>7.590909090909091, :standard_deviation=>5.770690236086651}}}
|
19
|
+
end
|
20
|
+
end
|
21
|
+
end
|
@@ -0,0 +1,68 @@
|
|
1
|
+
require 'engtagger'
|
2
|
+
|
3
|
+
module KnownItemSearchClassifier
|
4
|
+
class FeatureExtractor
|
5
|
+
def initialize string
|
6
|
+
@string = string
|
7
|
+
tagger = EngTagger.new
|
8
|
+
@tagged = tagger.get_readable string
|
9
|
+
@num_words = @tagged.scan(/\/[A-Z]{2}/).size.to_f
|
10
|
+
|
11
|
+
@mixed_case = is_mixed_case?
|
12
|
+
@punctuation_ratio = punctuation_ratio
|
13
|
+
@determiner_ratio = determiner_ratio
|
14
|
+
@proper_noun_ratio = proper_noun_ratio
|
15
|
+
|
16
|
+
#@num_keywords = count_keywords
|
17
|
+
#@refers_to_an_item_that_is_known = check_against_known_titles
|
18
|
+
|
19
|
+
end
|
20
|
+
def feature_array
|
21
|
+
return [@mixed_case, @punctuation_ratio, @determiner_ratio, @proper_noun_ratio, @num_words]
|
22
|
+
end
|
23
|
+
private
|
24
|
+
def is_mixed_case?
|
25
|
+
if @string =~ /[A-Z]/ and @string =~ /[a-z]/
|
26
|
+
return 1.0
|
27
|
+
end
|
28
|
+
return 0.0
|
29
|
+
end
|
30
|
+
def punctuation_ratio
|
31
|
+
num_punct = @tagged.scan(/\/PP/).size.to_f
|
32
|
+
return num_punct / @num_words
|
33
|
+
end
|
34
|
+
def determiner_ratio
|
35
|
+
num_det = @tagged.scan(/\/DET/).size.to_f
|
36
|
+
return num_det / @num_words
|
37
|
+
end
|
38
|
+
def proper_noun_ratio
|
39
|
+
num_prop_noun = @tagged.scan(/\/NNP/).size.to_f
|
40
|
+
return num_prop_noun / @num_words
|
41
|
+
end
|
42
|
+
def count_keywords
|
43
|
+
end
|
44
|
+
def check_against_known_titles
|
45
|
+
end
|
46
|
+
def count_keywords
|
47
|
+
keywords_to_match = ['journal', 'course', 'textbook']
|
48
|
+
num_keywords = 0
|
49
|
+
@query_string.split.each do |word|
|
50
|
+
if keywords_to_match.include? word.gsub(/[[:punct:]]/, '').downcase
|
51
|
+
num_keywords = num_keywords + 1
|
52
|
+
end
|
53
|
+
end
|
54
|
+
return num_keywords
|
55
|
+
end
|
56
|
+
def check_against_known_titles
|
57
|
+
known_titles = [
|
58
|
+
'salt sugar fat',
|
59
|
+
]
|
60
|
+
if known_titles.include? @query_string.downcase
|
61
|
+
return true
|
62
|
+
else
|
63
|
+
return false
|
64
|
+
end
|
65
|
+
end
|
66
|
+
end
|
67
|
+
end
|
68
|
+
|
@@ -0,0 +1,62 @@
|
|
1
|
+
require 'coveralls'
|
2
|
+
Coveralls.wear!
|
3
|
+
require 'minitest/autorun'
|
4
|
+
require './lib/known_item_search_classifier'
|
5
|
+
|
6
|
+
|
7
|
+
class KnownItemSearchClassifierTest < Minitest::Test
|
8
|
+
classifier = KnownItemSearchClassifier::Classifier.new
|
9
|
+
|
10
|
+
known_item_training_set = [
|
11
|
+
"little house on the",
|
12
|
+
"the inconvenient truth",
|
13
|
+
"the question of animal Culture by Kevin N Laland; Bennett G Galef ",
|
14
|
+
"Robinson Ken. Creative Schools: The Grassroots Revolution That’s Transforming Eduction. Viking. 2015. Print",
|
15
|
+
"The Boy in Zaquitos",
|
16
|
+
"The Mis-Education of the Negro",
|
17
|
+
"human relations interpersonal job-oriented skills",
|
18
|
+
"Research Methods for Business: A Skill-Building Approach Effectiveness of Instruction Performed through Computer-Assisted Activity Schedules on On-Schedule and Role-Play Skills of Children with Autism Spectrum Disorder",
|
19
|
+
"competency skills for the dental assiostant",
|
20
|
+
"Why did they kill?: Cambodia in the shadow of genocide",
|
21
|
+
"salt sugar fat",
|
22
|
+
"Making a Killing: Femicide, Free Trade, and La Frontera",
|
23
|
+
]
|
24
|
+
known_item_training_set.each do |query|
|
25
|
+
cleaned_up_query = query.gsub(/[[:punct:]]/, '').gsub(/[[:space:]]/, '_')
|
26
|
+
define_method("test_#{cleaned_up_query}_is_classified_as_known_item") do
|
27
|
+
assert_equal(:known, classifier.is_known_item_search?(query))
|
28
|
+
end
|
29
|
+
end
|
30
|
+
|
31
|
+
unknown_item_training_set = [
|
32
|
+
"earthworms",
|
33
|
+
"network security",
|
34
|
+
"work stress",
|
35
|
+
"mummies",
|
36
|
+
"benefits of eating healthyhy",
|
37
|
+
"benefits of eating healthy",
|
38
|
+
"megadosing vitamin c",
|
39
|
+
"nutrition",
|
40
|
+
"penquin",
|
41
|
+
"bananas",
|
42
|
+
"food sourcing",
|
43
|
+
"whey protein",
|
44
|
+
"exotic animals",
|
45
|
+
"sweet home oregon",
|
46
|
+
"taylor swift",
|
47
|
+
"catholicism",
|
48
|
+
"Professional baking ",
|
49
|
+
"concussions after the nfl",
|
50
|
+
"IVF the US",
|
51
|
+
"adoption children the US",
|
52
|
+
"Films for the hearing impaired",
|
53
|
+
"wolves and the ecosystem",
|
54
|
+
"dr. martin luther king",
|
55
|
+
]
|
56
|
+
unknown_item_training_set.each do |query|
|
57
|
+
cleaned_up_query = query.gsub(/[[:punct:]]/, '').gsub(/[[:space:]]/, '_')
|
58
|
+
define_method("test_#{cleaned_up_query}_is_not_false_positive") do
|
59
|
+
assert_equal(:unknown, classifier.is_known_item_search?(query))
|
60
|
+
end
|
61
|
+
end
|
62
|
+
end
|
metadata
ADDED
@@ -0,0 +1,106 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: known_item_search_classifier
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.1.0
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- Jane Sandberg
|
8
|
+
autorequire:
|
9
|
+
bindir: bin
|
10
|
+
cert_chain: []
|
11
|
+
date: 2016-11-13 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
name: engtagger
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - ">="
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: 0.2.1
|
20
|
+
type: :runtime
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - ">="
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: 0.2.1
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: gaussian_naive_bayes
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - ">="
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: 0.1.1
|
34
|
+
type: :runtime
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - ">="
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: 0.1.1
|
41
|
+
- !ruby/object:Gem::Dependency
|
42
|
+
name: minitest
|
43
|
+
requirement: !ruby/object:Gem::Requirement
|
44
|
+
requirements:
|
45
|
+
- - ">="
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
version: '0'
|
48
|
+
type: :development
|
49
|
+
prerelease: false
|
50
|
+
version_requirements: !ruby/object:Gem::Requirement
|
51
|
+
requirements:
|
52
|
+
- - ">="
|
53
|
+
- !ruby/object:Gem::Version
|
54
|
+
version: '0'
|
55
|
+
- !ruby/object:Gem::Dependency
|
56
|
+
name: coveralls
|
57
|
+
requirement: !ruby/object:Gem::Requirement
|
58
|
+
requirements:
|
59
|
+
- - '='
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: 0.7.0
|
62
|
+
type: :development
|
63
|
+
prerelease: false
|
64
|
+
version_requirements: !ruby/object:Gem::Requirement
|
65
|
+
requirements:
|
66
|
+
- - '='
|
67
|
+
- !ruby/object:Gem::Version
|
68
|
+
version: 0.7.0
|
69
|
+
description: Classify search query strings
|
70
|
+
email: sandbej@linnbenton.edu
|
71
|
+
executables: []
|
72
|
+
extensions: []
|
73
|
+
extra_rdoc_files: []
|
74
|
+
files:
|
75
|
+
- lib/known_item_search_classifier.rb
|
76
|
+
- lib/known_item_search_classifier/classifier.rb
|
77
|
+
- lib/known_item_search_classifier/default_training_set.rb
|
78
|
+
- lib/known_item_search_classifier/feature_extractor.rb
|
79
|
+
- test/known_item_search_classifier_test.rb
|
80
|
+
homepage: https://github.com/sandbergja/known_item_search_classifier
|
81
|
+
licenses:
|
82
|
+
- MIT
|
83
|
+
metadata: {}
|
84
|
+
post_install_message:
|
85
|
+
rdoc_options: []
|
86
|
+
require_paths:
|
87
|
+
- lib
|
88
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
89
|
+
requirements:
|
90
|
+
- - ">="
|
91
|
+
- !ruby/object:Gem::Version
|
92
|
+
version: '0'
|
93
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
94
|
+
requirements:
|
95
|
+
- - ">="
|
96
|
+
- !ruby/object:Gem::Version
|
97
|
+
version: '0'
|
98
|
+
requirements: []
|
99
|
+
rubyforge_project:
|
100
|
+
rubygems_version: 2.5.1
|
101
|
+
signing_key:
|
102
|
+
specification_version: 4
|
103
|
+
summary: A ruby gem that classifies search query strings as either known-item searches
|
104
|
+
or unknown-item searches
|
105
|
+
test_files:
|
106
|
+
- test/known_item_search_classifier_test.rb
|