wikipedia-vandalism_detection 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/.gitignore +17 -0
- data/Gemfile +6 -0
- data/LICENSE.txt +4 -0
- data/README.md +265 -0
- data/Rakefile +12 -0
- data/lib/java/LibSVM.jar +0 -0
- data/lib/java/SMOTE.jar +0 -0
- data/lib/java/balancedRandomForest.jar +0 -0
- data/lib/java/diffutils-1.3.0.jar +0 -0
- data/lib/java/libsvm.jar +0 -0
- data/lib/java/oneClassClassifier.jar +0 -0
- data/lib/java/realAdaBoost.jar +0 -0
- data/lib/java/swc-engine-1.1.0-jar-with-dependencies.jar +0 -0
- data/lib/java/sweble-wikitext-extractor.jar +0 -0
- data/lib/weka/classifiers/functions/lib_svm.rb +15 -0
- data/lib/weka/classifiers/meta/one_class_classifier.rb +25 -0
- data/lib/weka/classifiers/meta/real_ada_boost.rb +17 -0
- data/lib/weka/classifiers/trees/balanced_random_forest.rb +18 -0
- data/lib/weka/filters/supervised/instance/smote.rb +22 -0
- data/lib/wikipedia.rb +51 -0
- data/lib/wikipedia/vandalism_detection.rb +30 -0
- data/lib/wikipedia/vandalism_detection/algorithms.rb +18 -0
- data/lib/wikipedia/vandalism_detection/algorithms/kullback_leibler_divergence.rb +69 -0
- data/lib/wikipedia/vandalism_detection/classifier.rb +186 -0
- data/lib/wikipedia/vandalism_detection/configuration.rb +321 -0
- data/lib/wikipedia/vandalism_detection/diff.rb +27 -0
- data/lib/wikipedia/vandalism_detection/edit.rb +75 -0
- data/lib/wikipedia/vandalism_detection/evaluator.rb +606 -0
- data/lib/wikipedia/vandalism_detection/exceptions.rb +40 -0
- data/lib/wikipedia/vandalism_detection/feature_calculator.rb +89 -0
- data/lib/wikipedia/vandalism_detection/features.rb +67 -0
- data/lib/wikipedia/vandalism_detection/features/all_wordlists_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/all_wordlists_impact.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/anonymity.rb +19 -0
- data/lib/wikipedia/vandalism_detection/features/anonymity_previous.rb +29 -0
- data/lib/wikipedia/vandalism_detection/features/article_size.rb +18 -0
- data/lib/wikipedia/vandalism_detection/features/bad_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/bad_impact.rb +21 -0
- data/lib/wikipedia/vandalism_detection/features/base.rb +54 -0
- data/lib/wikipedia/vandalism_detection/features/biased_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/biased_impact.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/blanking.rb +25 -0
- data/lib/wikipedia/vandalism_detection/features/character_diversity.rb +25 -0
- data/lib/wikipedia/vandalism_detection/features/character_sequence.rb +19 -0
- data/lib/wikipedia/vandalism_detection/features/comment_bad_frequency.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/comment_biased_frequency.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/comment_length.rb +17 -0
- data/lib/wikipedia/vandalism_detection/features/comment_markup_frequency.rb +27 -0
- data/lib/wikipedia/vandalism_detection/features/comment_pronoun_frequency.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/comment_sex_frequency.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/comment_vulgarism_frequency.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/compressibility.rb +27 -0
- data/lib/wikipedia/vandalism_detection/features/contains_base.rb +18 -0
- data/lib/wikipedia/vandalism_detection/features/copyedit.rb +18 -0
- data/lib/wikipedia/vandalism_detection/features/digit_ratio.rb +24 -0
- data/lib/wikipedia/vandalism_detection/features/edits_per_user.rb +65 -0
- data/lib/wikipedia/vandalism_detection/features/emoticons_frequency.rb +27 -0
- data/lib/wikipedia/vandalism_detection/features/emoticons_impact.rb +29 -0
- data/lib/wikipedia/vandalism_detection/features/frequency_base.rb +20 -0
- data/lib/wikipedia/vandalism_detection/features/impact_base.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/inserted_character_distribution.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/inserted_external_links.rb +18 -0
- data/lib/wikipedia/vandalism_detection/features/inserted_internal_links.rb +18 -0
- data/lib/wikipedia/vandalism_detection/features/inserted_size.rb +20 -0
- data/lib/wikipedia/vandalism_detection/features/inserted_words.rb +18 -0
- data/lib/wikipedia/vandalism_detection/features/longest_word.rb +20 -0
- data/lib/wikipedia/vandalism_detection/features/markup_frequency.rb +27 -0
- data/lib/wikipedia/vandalism_detection/features/markup_impact.rb +27 -0
- data/lib/wikipedia/vandalism_detection/features/non_alphanumeric_ratio.rb +24 -0
- data/lib/wikipedia/vandalism_detection/features/personal_life.rb +18 -0
- data/lib/wikipedia/vandalism_detection/features/pronoun_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/pronoun_impact.rb +21 -0
- data/lib/wikipedia/vandalism_detection/features/removed_all_wordlists_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/removed_bad_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/removed_biased_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/removed_character_distribution.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/removed_emoticons_frequency.rb +27 -0
- data/lib/wikipedia/vandalism_detection/features/removed_markup_frequency.rb +28 -0
- data/lib/wikipedia/vandalism_detection/features/removed_pronoun_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/removed_sex_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/removed_size.rb +20 -0
- data/lib/wikipedia/vandalism_detection/features/removed_vulgarism_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/removed_words.rb +18 -0
- data/lib/wikipedia/vandalism_detection/features/replacement_similarity.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/reverted.rb +18 -0
- data/lib/wikipedia/vandalism_detection/features/revisions_character_distribution.rb +21 -0
- data/lib/wikipedia/vandalism_detection/features/same_editor.rb +29 -0
- data/lib/wikipedia/vandalism_detection/features/sex_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/sex_impact.rb +21 -0
- data/lib/wikipedia/vandalism_detection/features/size_increment.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/size_ratio.rb +26 -0
- data/lib/wikipedia/vandalism_detection/features/term_frequency.rb +25 -0
- data/lib/wikipedia/vandalism_detection/features/time_interval.rb +31 -0
- data/lib/wikipedia/vandalism_detection/features/time_of_day.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/upper_case_ratio.rb +24 -0
- data/lib/wikipedia/vandalism_detection/features/upper_case_words_ratio.rb +31 -0
- data/lib/wikipedia/vandalism_detection/features/upper_to_lower_case_ratio.rb +24 -0
- data/lib/wikipedia/vandalism_detection/features/user_reputation.rb +38 -0
- data/lib/wikipedia/vandalism_detection/features/vulgarism_frequency.rb +23 -0
- data/lib/wikipedia/vandalism_detection/features/vulgarism_impact.rb +22 -0
- data/lib/wikipedia/vandalism_detection/features/weekday.rb +21 -0
- data/lib/wikipedia/vandalism_detection/features/words_increment.rb +22 -0
- data/lib/wikipedia/vandalism_detection/instances.rb +130 -0
- data/lib/wikipedia/vandalism_detection/page.rb +88 -0
- data/lib/wikipedia/vandalism_detection/page_parser.rb +52 -0
- data/lib/wikipedia/vandalism_detection/revision.rb +69 -0
- data/lib/wikipedia/vandalism_detection/revision_parser.rb +43 -0
- data/lib/wikipedia/vandalism_detection/test_dataset.rb +367 -0
- data/lib/wikipedia/vandalism_detection/text.rb +18 -0
- data/lib/wikipedia/vandalism_detection/training_dataset.rb +303 -0
- data/lib/wikipedia/vandalism_detection/version.rb +5 -0
- data/lib/wikipedia/vandalism_detection/wikitext_extractor.rb +80 -0
- data/lib/wikipedia/vandalism_detection/word_lists.rb +19 -0
- data/lib/wikipedia/vandalism_detection/word_lists/bad.rb +12 -0
- data/lib/wikipedia/vandalism_detection/word_lists/biased.rb +21 -0
- data/lib/wikipedia/vandalism_detection/word_lists/emoticons.rb +22 -0
- data/lib/wikipedia/vandalism_detection/word_lists/markup.rb +12 -0
- data/lib/wikipedia/vandalism_detection/word_lists/pronouns.rb +15 -0
- data/lib/wikipedia/vandalism_detection/word_lists/sex.rb +12 -0
- data/lib/wikipedia/vandalism_detection/word_lists/vulgarism.rb +97 -0
- data/spec/factories/edit.rb +20 -0
- data/spec/factories/page.rb +13 -0
- data/spec/factories/revision.rb +51 -0
- data/spec/resources/config/config.yml +35 -0
- data/spec/resources/corpora/test/edits.csv +8 -0
- data/spec/resources/corpora/test/ground-truth.txt +3 -0
- data/spec/resources/corpora/test/revisions/part-1/326471754.txt +199 -0
- data/spec/resources/corpora/test/revisions/part-1/326873205.txt +137 -0
- data/spec/resources/corpora/test/revisions/part-1/328774035.txt +162 -0
- data/spec/resources/corpora/test/revisions/part-2/307084144.txt +137 -0
- data/spec/resources/corpora/test/revisions/part-2/326978767.txt +199 -0
- data/spec/resources/corpora/test/revisions/part-2/328774110.txt +162 -0
- data/spec/resources/corpora/training/annotations.csv +7 -0
- data/spec/resources/corpora/training/edits.csv +7 -0
- data/spec/resources/corpora/training/revisions/part-1/326471754.txt +199 -0
- data/spec/resources/corpora/training/revisions/part-1/326873205.txt +137 -0
- data/spec/resources/corpora/training/revisions/part-1/328774035.txt +162 -0
- data/spec/resources/corpora/training/revisions/part-2/307084144.txt +137 -0
- data/spec/resources/corpora/training/revisions/part-2/326978767.txt +199 -0
- data/spec/resources/corpora/training/revisions/part-2/328774110.txt +162 -0
- data/spec/resources/page_with_redirects.xml +85 -0
- data/spec/resources/redirect_page.xml +59 -0
- data/spec/resources/revision_simplified.xml +13 -0
- data/spec/resources/sample_revision.txt +137 -0
- data/spec/resources/sample_revision_clean_text.txt +1 -0
- data/spec/resources/sample_revision_plain_text.txt +183 -0
- data/spec/resources/vandalism_on_wikipedia.xml +234 -0
- data/spec/resources/vandalism_on_wikipedia_simplified.xml +119 -0
- data/spec/resources/wikipedia_tokens.txt +30 -0
- data/spec/spec_helper.rb +22 -0
- data/spec/support/macros/file_reading.rb +7 -0
- data/spec/support/macros/test_configuration.rb +71 -0
- data/spec/vandalism_detection/algorithms/kullback_leibler_divergence_spec.rb +36 -0
- data/spec/vandalism_detection/classifier_spec.rb +317 -0
- data/spec/vandalism_detection/configuration_spec.rb +517 -0
- data/spec/vandalism_detection/diff_spec.rb +40 -0
- data/spec/vandalism_detection/edit_spec.rb +137 -0
- data/spec/vandalism_detection/evaluator_spec.rb +671 -0
- data/spec/vandalism_detection/feature_calculator_spec.rb +128 -0
- data/spec/vandalism_detection/features/all_wordlists_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/all_wordlists_impact_spec.rb +58 -0
- data/spec/vandalism_detection/features/anonymity_previous_spec.rb +61 -0
- data/spec/vandalism_detection/features/anonymity_spec.rb +23 -0
- data/spec/vandalism_detection/features/article_size_spec.rb +35 -0
- data/spec/vandalism_detection/features/bad_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/bad_impact_spec.rb +59 -0
- data/spec/vandalism_detection/features/base_spec.rb +49 -0
- data/spec/vandalism_detection/features/biased_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/biased_impact_spec.rb +58 -0
- data/spec/vandalism_detection/features/blanking_spec.rb +38 -0
- data/spec/vandalism_detection/features/character_diversity_spec.rb +35 -0
- data/spec/vandalism_detection/features/character_sequence_spec.rb +37 -0
- data/spec/vandalism_detection/features/comment_bad_frequency_spec.rb +34 -0
- data/spec/vandalism_detection/features/comment_biased_frequency_spec.rb +34 -0
- data/spec/vandalism_detection/features/comment_length_spec.rb +27 -0
- data/spec/vandalism_detection/features/comment_markup_frequency_spec.rb +34 -0
- data/spec/vandalism_detection/features/comment_pronoun_frequency_spec.rb +34 -0
- data/spec/vandalism_detection/features/comment_sex_frequency_spec.rb +34 -0
- data/spec/vandalism_detection/features/comment_vulgarism_frequency_spec.rb +34 -0
- data/spec/vandalism_detection/features/compressibility_spec.rb +42 -0
- data/spec/vandalism_detection/features/contains_base_spec.rb +33 -0
- data/spec/vandalism_detection/features/copyedit_spec.rb +33 -0
- data/spec/vandalism_detection/features/digit_ratio_spec.rb +35 -0
- data/spec/vandalism_detection/features/edits_per_user_spec.rb +49 -0
- data/spec/vandalism_detection/features/emoticons_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/emoticons_impact_spec.rb +51 -0
- data/spec/vandalism_detection/features/frequency_base_spec.rb +26 -0
- data/spec/vandalism_detection/features/impact_base_spec.rb +41 -0
- data/spec/vandalism_detection/features/inserted_character_distribution_spec.rb +46 -0
- data/spec/vandalism_detection/features/inserted_external_links_spec.rb +35 -0
- data/spec/vandalism_detection/features/inserted_internal_links_spec.rb +35 -0
- data/spec/vandalism_detection/features/inserted_size_spec.rb +35 -0
- data/spec/vandalism_detection/features/inserted_words_spec.rb +35 -0
- data/spec/vandalism_detection/features/longest_word_spec.rb +35 -0
- data/spec/vandalism_detection/features/markup_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/markup_impact_spec.rb +59 -0
- data/spec/vandalism_detection/features/non_alphanumeric_ratio_spec.rb +35 -0
- data/spec/vandalism_detection/features/personal_life_spec.rb +26 -0
- data/spec/vandalism_detection/features/pronoun_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/pronoun_impact_spec.rb +59 -0
- data/spec/vandalism_detection/features/removed_all_wordlists_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/removed_bad_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/removed_biased_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/removed_character_distribution_spec.rb +46 -0
- data/spec/vandalism_detection/features/removed_emoticons_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/removed_markup_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/removed_pronoun_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/removed_sex_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/removed_size_spec.rb +35 -0
- data/spec/vandalism_detection/features/removed_vulgarism_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/removed_words_spec.rb +35 -0
- data/spec/vandalism_detection/features/replacement_similarity_spec.rb +44 -0
- data/spec/vandalism_detection/features/reverted_spec.rb +28 -0
- data/spec/vandalism_detection/features/revisions_character_distribution_spec.rb +46 -0
- data/spec/vandalism_detection/features/same_editor_spec.rb +60 -0
- data/spec/vandalism_detection/features/sex_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/sex_impact_spec.rb +59 -0
- data/spec/vandalism_detection/features/size_increment_spec.rb +35 -0
- data/spec/vandalism_detection/features/size_ratio_spec.rb +57 -0
- data/spec/vandalism_detection/features/term_frequency_spec.rb +38 -0
- data/spec/vandalism_detection/features/time_interval_spec.rb +50 -0
- data/spec/vandalism_detection/features/time_of_day_spec.rb +22 -0
- data/spec/vandalism_detection/features/upper_case_ratio_spec.rb +35 -0
- data/spec/vandalism_detection/features/upper_case_words_ratio_spec.rb +37 -0
- data/spec/vandalism_detection/features/upper_to_lower_case_ratio_spec.rb +35 -0
- data/spec/vandalism_detection/features/user_reputation_spec.rb +52 -0
- data/spec/vandalism_detection/features/vulgarism_frequency_spec.rb +36 -0
- data/spec/vandalism_detection/features/vulgarism_impact_spec.rb +58 -0
- data/spec/vandalism_detection/features/weekday_spec.rb +22 -0
- data/spec/vandalism_detection/features/words_increment_spec.rb +35 -0
- data/spec/vandalism_detection/instances_spec.rb +156 -0
- data/spec/vandalism_detection/page_parser_spec.rb +184 -0
- data/spec/vandalism_detection/page_spec.rb +135 -0
- data/spec/vandalism_detection/revision_parser_spec.rb +53 -0
- data/spec/vandalism_detection/revision_spec.rb +115 -0
- data/spec/vandalism_detection/test_dataset_spec.rb +231 -0
- data/spec/vandalism_detection/text_spec.rb +29 -0
- data/spec/vandalism_detection/training_dataset_spec.rb +264 -0
- data/spec/vandalism_detection/wikitext_extractor_spec.rb +72 -0
- data/spec/weka/classifiers/functions/lib_svm_spec.rb +38 -0
- data/spec/weka/classifiers/meta/one_class_classifier_spec.rb +76 -0
- data/spec/weka/classifiers/meta/real_ada_boost_spec.rb +40 -0
- data/spec/weka/classifiers/trees/balanced_random_forest_spec.rb +40 -0
- data/spec/weka/filters/supervised/instance/smote_spec.rb +6 -0
- data/wikipedia-vandalism_detection.gemspec +30 -0
- metadata +512 -0
@@ -0,0 +1,42 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
require 'zlib'
|
3
|
+
|
4
|
+
describe Wikipedia::VandalismDetection::Features::Compressibility do
|
5
|
+
|
6
|
+
before do
|
7
|
+
@feature = Wikipedia::VandalismDetection::Features::Compressibility.new
|
8
|
+
end
|
9
|
+
|
10
|
+
it { should be_a Wikipedia::VandalismDetection::Features::Base }
|
11
|
+
|
12
|
+
describe "#calculate" do
|
13
|
+
|
14
|
+
it "returns the ratio of compressed text size to uncompressed text size" do
|
15
|
+
old_text = 'text'
|
16
|
+
new_text = 'text [[If this is a quite long textpart]] of normal words then it might be less possible to be a vandalism.'
|
17
|
+
|
18
|
+
old_revision = build(:old_revision, text: Wikipedia::VandalismDetection::Text.new(old_text))
|
19
|
+
new_revision = build(:new_revision, text: Wikipedia::VandalismDetection::Text.new(new_text))
|
20
|
+
edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
|
21
|
+
|
22
|
+
bytesize = 10.0
|
23
|
+
|
24
|
+
Zlib::Deflate.stub(deflate: Wikipedia::VandalismDetection::Text.new)
|
25
|
+
Wikipedia::VandalismDetection::Text.any_instance.stub(bytesize: bytesize)
|
26
|
+
ratio = bytesize / (bytesize + bytesize)
|
27
|
+
|
28
|
+
expect(@feature.calculate(edit)).to eq ratio
|
29
|
+
end
|
30
|
+
|
31
|
+
it "returns 0.5 on emtpy inserted text" do
|
32
|
+
old_text = Wikipedia::VandalismDetection::Text.new("deletion text")
|
33
|
+
new_text = Wikipedia::VandalismDetection::Text.new(" text")
|
34
|
+
|
35
|
+
old_revision = build(:old_revision, text: old_text)
|
36
|
+
new_revision = build(:new_revision, text: new_text)
|
37
|
+
edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
|
38
|
+
|
39
|
+
expect(@feature.calculate(edit)).to eq 0.5
|
40
|
+
end
|
41
|
+
end
|
42
|
+
end
|
@@ -0,0 +1,33 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
describe Wikipedia::VandalismDetection::Features::ContainsBase do
|
4
|
+
|
5
|
+
before do
|
6
|
+
@base = Wikipedia::VandalismDetection::Features::ContainsBase.new
|
7
|
+
end
|
8
|
+
|
9
|
+
it { should be_a Wikipedia::VandalismDetection::Features::Base }
|
10
|
+
|
11
|
+
describe "#contains" do
|
12
|
+
|
13
|
+
it "returns 1 if a given text contains the given terms array" do
|
14
|
+
text = "Content including text"
|
15
|
+
expect(@base.contains(text, ['content', 'anything'])).to eq 1
|
16
|
+
end
|
17
|
+
|
18
|
+
it "returns 1 if a given text contains the given string" do
|
19
|
+
text = "Content including text"
|
20
|
+
expect(@base.contains(text, 'content')).to eq 1
|
21
|
+
end
|
22
|
+
|
23
|
+
it "returns 0 if a given text does not contain the given string" do
|
24
|
+
text = "not containing anything con tent"
|
25
|
+
expect(@base.contains(text, 'content')).to eq 0
|
26
|
+
end
|
27
|
+
|
28
|
+
it "returns 0 if a given text does not contain any of the given terms" do
|
29
|
+
text = "not containing anything con tent"
|
30
|
+
expect(@base.contains(text, ['content', 'text'])).to eq 0
|
31
|
+
end
|
32
|
+
end
|
33
|
+
end
|
@@ -0,0 +1,33 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
describe Wikipedia::VandalismDetection::Features::Copyedit do
|
4
|
+
|
5
|
+
before do
|
6
|
+
@feature = Wikipedia::VandalismDetection::Features::Copyedit.new
|
7
|
+
end
|
8
|
+
|
9
|
+
it { should be_a Wikipedia::VandalismDetection::Features::ContainsBase }
|
10
|
+
|
11
|
+
describe "#calculate" do
|
12
|
+
|
13
|
+
it "returns 1 if the edit comment includes 'copyedit'" do
|
14
|
+
comment = Wikipedia::VandalismDetection::Text.new "copyediting content"
|
15
|
+
edit = build :edit, new_revision: build(:new_revision, comment: comment)
|
16
|
+
|
17
|
+
expect(@feature.calculate(edit)).to eq 1
|
18
|
+
end
|
19
|
+
|
20
|
+
it "returns 1 if the edit comment includes 'copy edit'" do
|
21
|
+
comment = Wikipedia::VandalismDetection::Text.new "copy editing content"
|
22
|
+
edit = build :edit, new_revision: build(:new_revision, comment: comment)
|
23
|
+
|
24
|
+
expect(@feature.calculate(edit)).to eq 1
|
25
|
+
end
|
26
|
+
|
27
|
+
it "returns 0 on emtpy comment" do
|
28
|
+
edit = build :edit, new_revision: build(:new_revision, comment: "")
|
29
|
+
|
30
|
+
expect(@feature.calculate(edit)).to eq 0
|
31
|
+
end
|
32
|
+
end
|
33
|
+
end
|
@@ -0,0 +1,35 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
describe Wikipedia::VandalismDetection::Features::DigitRatio do
|
4
|
+
|
5
|
+
before do
|
6
|
+
@feature = Wikipedia::VandalismDetection::Features::DigitRatio.new
|
7
|
+
end
|
8
|
+
|
9
|
+
it { should be_a Wikipedia::VandalismDetection::Features::Base }
|
10
|
+
|
11
|
+
describe "#calculate" do
|
12
|
+
|
13
|
+
it "returns the digit to all letters ratio of the edit's new revision inserted text" do
|
14
|
+
old_text = Wikipedia::VandalismDetection::Text.new('text1')
|
15
|
+
new_text = Wikipedia::VandalismDetection::Text.new 'text1 [[1A4 B6 8Cd]]' # 3 digit letters of total 8 letters
|
16
|
+
|
17
|
+
old_revision = build(:old_revision, text: old_text)
|
18
|
+
new_revision = build(:new_revision, text: new_text)
|
19
|
+
edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
|
20
|
+
|
21
|
+
expect(@feature.calculate(edit)).to eq (1.0 + 4) / (1.0 + 8)
|
22
|
+
end
|
23
|
+
|
24
|
+
it "returns 0.0 if no text inserted" do
|
25
|
+
old_text = Wikipedia::VandalismDetection::Text.new("deletion text")
|
26
|
+
new_text = Wikipedia::VandalismDetection::Text.new("text")
|
27
|
+
|
28
|
+
old_revision = build(:old_revision, text: old_text)
|
29
|
+
new_revision = build(:new_revision, text: new_text)
|
30
|
+
edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
|
31
|
+
|
32
|
+
expect(@feature.calculate(edit)).to eq 0.0
|
33
|
+
end
|
34
|
+
end
|
35
|
+
end
|
@@ -0,0 +1,49 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
describe Wikipedia::VandalismDetection::Features::EditsPerUser do
|
4
|
+
|
5
|
+
before do
|
6
|
+
@feature = Wikipedia::VandalismDetection::Features::EditsPerUser.new
|
7
|
+
end
|
8
|
+
|
9
|
+
it { should be_a Wikipedia::VandalismDetection::Features::Base }
|
10
|
+
|
11
|
+
describe "#calculate" do
|
12
|
+
|
13
|
+
describe "online" do
|
14
|
+
it "returns the number of previously submitted edit from the same IP or ID" do
|
15
|
+
#http://en.wikipedia.org/w/api.php?action=query&format=json&list=usercontribs&ucuser=<name or ip>&ucprop=ids
|
16
|
+
old_revision = build(:old_revision, id: '527136737')
|
17
|
+
new_revision = build(:new_revision, id: '527137015', parent_id: '527136737',
|
18
|
+
contributor: '142.11.81.219', timestamp: '2012-12-09T05:30:07Z' )
|
19
|
+
|
20
|
+
edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
|
21
|
+
|
22
|
+
expect(@feature.calculate(edit)).to eq 1
|
23
|
+
end
|
24
|
+
end
|
25
|
+
|
26
|
+
describe "offline" do
|
27
|
+
before do
|
28
|
+
page = build(:page)
|
29
|
+
page.id = '1234'
|
30
|
+
page.title = 'Page Title'
|
31
|
+
|
32
|
+
# contributor: see factories/page.rb !
|
33
|
+
old_revision = build(:new_revision, contributor: 'User')
|
34
|
+
new_revision = build(:even_newer_revision, contributor: 'User')
|
35
|
+
|
36
|
+
@edit = build(:edit, old_revision: old_revision, new_revision: new_revision, page: page)
|
37
|
+
end
|
38
|
+
|
39
|
+
it "does not use an API call if the edit has a page reference" do
|
40
|
+
expect(Wikipedia).to_not receive :api_request
|
41
|
+
@feature.calculate(@edit)
|
42
|
+
end
|
43
|
+
|
44
|
+
it "returns the number of previously submitted edit from the same IP or ID" do
|
45
|
+
expect(@feature.calculate(@edit)).to eq 1
|
46
|
+
end
|
47
|
+
end
|
48
|
+
end
|
49
|
+
end
|
@@ -0,0 +1,36 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
describe Wikipedia::VandalismDetection::Features::EmoticonsFrequency do
|
4
|
+
|
5
|
+
before do
|
6
|
+
@feature = Wikipedia::VandalismDetection::Features::EmoticonsFrequency.new
|
7
|
+
end
|
8
|
+
|
9
|
+
it { should be_a Wikipedia::VandalismDetection::Features::Base }
|
10
|
+
|
11
|
+
describe "#calculate" do
|
12
|
+
|
13
|
+
it "returns the number of emoticons relative to all words count" do
|
14
|
+
# total 10 words, 3 emoticons
|
15
|
+
old_text = Wikipedia::VandalismDetection::Text.new 'Old :-).'
|
16
|
+
new_text = Wikipedia::VandalismDetection::Text.new "Old :-). ;) you do love icons and emoticons? :D :P, yeah."
|
17
|
+
|
18
|
+
old_revision = build(:old_revision, text: old_text)
|
19
|
+
new_revision = build(:new_revision, text: new_text)
|
20
|
+
edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
|
21
|
+
|
22
|
+
expect(@feature.calculate(edit)).to eq 3.0 / 10.0
|
23
|
+
end
|
24
|
+
|
25
|
+
it "returns 0.0 on emtpy clean text revisions" do
|
26
|
+
old_text = Wikipedia::VandalismDetection::Text.new 'Old :-).'
|
27
|
+
new_text = Wikipedia::VandalismDetection::Text.new "Old :-). {{speedy deletion}}"
|
28
|
+
|
29
|
+
old_revision = build(:old_revision, text: old_text)
|
30
|
+
new_revision = build(:new_revision, text: new_text)
|
31
|
+
edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
|
32
|
+
|
33
|
+
expect(@feature.calculate(edit)).to eq 0.0
|
34
|
+
end
|
35
|
+
end
|
36
|
+
end
|
@@ -0,0 +1,51 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
describe Wikipedia::VandalismDetection::Features::EmoticonsImpact do
|
4
|
+
|
5
|
+
before do
|
6
|
+
@feature = Wikipedia::VandalismDetection::Features::EmoticonsImpact.new
|
7
|
+
end
|
8
|
+
|
9
|
+
it { should be_a Wikipedia::VandalismDetection::Features::Base }
|
10
|
+
|
11
|
+
describe "#calculate" do
|
12
|
+
|
13
|
+
it "returns the impact of emoticon words of the edit's new revision text" do
|
14
|
+
# 3 emoticons
|
15
|
+
old_text = ":) Hi you I got some :-X, you know ;)"
|
16
|
+
|
17
|
+
# 4 emoticons
|
18
|
+
new_text = ":) Hi (=you) I added another :-X you know ;)? (="
|
19
|
+
|
20
|
+
old_revision = build(:old_revision, text: old_text)
|
21
|
+
new_revision = build(:new_revision, text: new_text)
|
22
|
+
edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
|
23
|
+
|
24
|
+
expect(@feature.calculate(edit)).to eq 3.0 / (3.0 + 4.0)
|
25
|
+
end
|
26
|
+
|
27
|
+
it "returns 0.5 on both no terms in text revisions" do
|
28
|
+
old_revision = build(:old_revision, text: "")
|
29
|
+
new_revision = build(:new_revision, text: "")
|
30
|
+
edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
|
31
|
+
|
32
|
+
expect(@feature.calculate(edit)).to eq 0.5
|
33
|
+
end
|
34
|
+
|
35
|
+
it "returns 0.0 on emtpy text of old revision" do
|
36
|
+
old_revision = build(:old_revision, text: "")
|
37
|
+
new_revision = build(:new_revision, text: ":)")
|
38
|
+
edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
|
39
|
+
|
40
|
+
expect(@feature.calculate(edit)).to eq 0.0
|
41
|
+
end
|
42
|
+
|
43
|
+
it "returns 1.0 on emtpy text of new revision" do
|
44
|
+
old_revision = build(:old_revision, text: ":)")
|
45
|
+
new_revision = build(:new_revision, text: "")
|
46
|
+
edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
|
47
|
+
|
48
|
+
expect(@feature.calculate(edit)).to eq 1.0
|
49
|
+
end
|
50
|
+
end
|
51
|
+
end
|
@@ -0,0 +1,26 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
require 'wikipedia/vandalism_detection/features/frequency_base'
|
3
|
+
|
4
|
+
describe Wikipedia::VandalismDetection::Features::FrequencyBase do
|
5
|
+
|
6
|
+
before do
|
7
|
+
@feature = Wikipedia::VandalismDetection::Features::FrequencyBase.new
|
8
|
+
@terms = Wikipedia::VandalismDetection::WordLists::PRONOUNS
|
9
|
+
end
|
10
|
+
|
11
|
+
it { should be_a Wikipedia::VandalismDetection::Features::Base }
|
12
|
+
|
13
|
+
describe "#frequency" do
|
14
|
+
|
15
|
+
it { should respond_to :frequency }
|
16
|
+
|
17
|
+
it "returns the frequency in percentage of given word counts" do
|
18
|
+
text = "I am, i like you."
|
19
|
+
expect(@feature.frequency(text , @terms)).to eq 3.0/5.0
|
20
|
+
end
|
21
|
+
|
22
|
+
it "returns 0.0 if total word count is zero" do
|
23
|
+
expect(@feature.frequency("", @terms)).to eq 0.0
|
24
|
+
end
|
25
|
+
end
|
26
|
+
end
|
@@ -0,0 +1,41 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
require 'wikipedia/vandalism_detection/features/impact_base'
|
3
|
+
|
4
|
+
describe Wikipedia::VandalismDetection::Features::ImpactBase do
|
5
|
+
|
6
|
+
before do
|
7
|
+
@feature = Wikipedia::VandalismDetection::Features::ImpactBase.new
|
8
|
+
@pronouns = Wikipedia::VandalismDetection::WordLists::PRONOUNS
|
9
|
+
end
|
10
|
+
|
11
|
+
it { should be_a Wikipedia::VandalismDetection::Features::Base }
|
12
|
+
|
13
|
+
describe "#impact" do
|
14
|
+
|
15
|
+
it { should respond_to :impact }
|
16
|
+
|
17
|
+
it "returns the impact in percentage of given terms in old realitve to new text" do
|
18
|
+
# 3 pronouns
|
19
|
+
old_text = "Your old text will be mine or Your's"
|
20
|
+
# 4 pronouns
|
21
|
+
new_text = "My new text and your old text will be ours and mine"
|
22
|
+
|
23
|
+
text = "I am, i like you."
|
24
|
+
expect(@feature.impact(old_text , new_text, @pronouns)).to eq 3.0 / (3.0 + 4.0)
|
25
|
+
end
|
26
|
+
|
27
|
+
it "returns 0.0 if old terms word count is zero" do
|
28
|
+
new_text = "My new text and your old text will be ours and mine"
|
29
|
+
expect(@feature.impact("", new_text, @pronouns)).to eq 0.0
|
30
|
+
end
|
31
|
+
|
32
|
+
it "returns 1.0 if new terms word count is zero" do
|
33
|
+
old_text = "My new text and your old text will be ours and mine"
|
34
|
+
expect(@feature.impact(old_text, "", @pronouns)).to eq 1.0
|
35
|
+
end
|
36
|
+
|
37
|
+
it "returns 0.5 if both terms word count is zero" do
|
38
|
+
expect(@feature.impact("", "", @pronouns)).to eq 0.5
|
39
|
+
end
|
40
|
+
end
|
41
|
+
end
|
@@ -0,0 +1,46 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
describe Wikipedia::VandalismDetection::Features::InsertedCharacterDistribution do
|
4
|
+
|
5
|
+
before do
|
6
|
+
@feature = Wikipedia::VandalismDetection::Features::InsertedCharacterDistribution.new
|
7
|
+
end
|
8
|
+
|
9
|
+
it { should be_a Wikipedia::VandalismDetection::Features::Base }
|
10
|
+
|
11
|
+
describe "#calculate" do
|
12
|
+
|
13
|
+
it "returns the Kullback-Leibler Divergence of the inserted text's character distribution" do
|
14
|
+
old_text = Wikipedia::VandalismDetection::Text.new('old text')
|
15
|
+
new_text = Wikipedia::VandalismDetection::Text.new 'old text [[new inserted text]] given dero 9'
|
16
|
+
|
17
|
+
old_revision = build(:old_revision, text: old_text)
|
18
|
+
new_revision = build(:new_revision, text: new_text)
|
19
|
+
edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
|
20
|
+
|
21
|
+
expect(@feature.calculate(edit)).to eq 1.6609633564650683
|
22
|
+
end
|
23
|
+
|
24
|
+
it "returns missing value if no alphanumeric characters are inserted" do
|
25
|
+
old_text = Wikipedia::VandalismDetection::Text.new("old text")
|
26
|
+
new_text = Wikipedia::VandalismDetection::Text.new("old text !* [[?]]")
|
27
|
+
|
28
|
+
old_revision = build(:old_revision, text: old_text)
|
29
|
+
new_revision = build(:new_revision, text: new_text)
|
30
|
+
edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
|
31
|
+
|
32
|
+
expect(@feature.calculate(edit)).to eq Wikipedia::VandalismDetection::Features::MISSING_VALUE
|
33
|
+
end
|
34
|
+
|
35
|
+
it "returns missing value if no text inserted" do
|
36
|
+
old_text = Wikipedia::VandalismDetection::Text.new("deletion text")
|
37
|
+
new_text = Wikipedia::VandalismDetection::Text.new("text")
|
38
|
+
|
39
|
+
old_revision = build(:old_revision, text: old_text)
|
40
|
+
new_revision = build(:new_revision, text: new_text)
|
41
|
+
edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
|
42
|
+
|
43
|
+
expect(@feature.calculate(edit)).to eq Wikipedia::VandalismDetection::Features::MISSING_VALUE
|
44
|
+
end
|
45
|
+
end
|
46
|
+
end
|
@@ -0,0 +1,35 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
describe Wikipedia::VandalismDetection::Features::InsertedExternalLinks do
|
4
|
+
|
5
|
+
before do
|
6
|
+
@feature = Wikipedia::VandalismDetection::Features::InsertedExternalLinks.new
|
7
|
+
end
|
8
|
+
|
9
|
+
it { should be_a Wikipedia::VandalismDetection::Features::Base }
|
10
|
+
|
11
|
+
describe "#calculate" do
|
12
|
+
|
13
|
+
it "returns the number of inserted external links" do
|
14
|
+
old_revision_text = Wikipedia::VandalismDetection::Text.new '123'
|
15
|
+
new_revision_text = Wikipedia::VandalismDetection::Text.new '123 [http://google.com Google] https://example.com'
|
16
|
+
|
17
|
+
old_revision = build(:old_revision, text: old_revision_text)
|
18
|
+
new_revision = build(:new_revision, text: new_revision_text)
|
19
|
+
edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
|
20
|
+
|
21
|
+
expect(@feature.calculate(edit)).to eq 2
|
22
|
+
end
|
23
|
+
|
24
|
+
it "returns 0 if no inserted text" do
|
25
|
+
old_revision_text = Wikipedia::VandalismDetection::Text.new '123 456789'
|
26
|
+
new_revision_text = Wikipedia::VandalismDetection::Text.new '123' # 0 inserted
|
27
|
+
|
28
|
+
old_revision = build(:old_revision, text: old_revision_text)
|
29
|
+
new_revision = build(:new_revision, text: new_revision_text)
|
30
|
+
edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
|
31
|
+
|
32
|
+
expect(@feature.calculate(edit)).to eq 0
|
33
|
+
end
|
34
|
+
end
|
35
|
+
end
|
@@ -0,0 +1,35 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
describe Wikipedia::VandalismDetection::Features::InsertedInternalLinks do
|
4
|
+
|
5
|
+
before do
|
6
|
+
@feature = Wikipedia::VandalismDetection::Features::InsertedInternalLinks.new
|
7
|
+
end
|
8
|
+
|
9
|
+
it { should be_a Wikipedia::VandalismDetection::Features::Base }
|
10
|
+
|
11
|
+
describe "#calculate" do
|
12
|
+
|
13
|
+
it "returns the number of inserted internal links" do
|
14
|
+
old_revision_text = Wikipedia::VandalismDetection::Text.new '123'
|
15
|
+
new_revision_text = Wikipedia::VandalismDetection::Text.new '123 [[link]] [[linkname|link]]'
|
16
|
+
|
17
|
+
old_revision = build(:old_revision, text: old_revision_text)
|
18
|
+
new_revision = build(:new_revision, text: new_revision_text)
|
19
|
+
edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
|
20
|
+
|
21
|
+
expect(@feature.calculate(edit)).to eq 2
|
22
|
+
end
|
23
|
+
|
24
|
+
it "returns 0 if no inserted text" do
|
25
|
+
old_revision_text = Wikipedia::VandalismDetection::Text.new '123 456789'
|
26
|
+
new_revision_text = Wikipedia::VandalismDetection::Text.new '123' # 0 inserted
|
27
|
+
|
28
|
+
old_revision = build(:old_revision, text: old_revision_text)
|
29
|
+
new_revision = build(:new_revision, text: new_revision_text)
|
30
|
+
edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
|
31
|
+
|
32
|
+
expect(@feature.calculate(edit)).to eq 0
|
33
|
+
end
|
34
|
+
end
|
35
|
+
end
|