wikipedia-vandalism_detection 0.0.1

Sign up to get free protection for your applications and to get access to all the features.
Files changed (247) hide show
  1. checksums.yaml +7 -0
  2. data/.gitignore +17 -0
  3. data/Gemfile +6 -0
  4. data/LICENSE.txt +4 -0
  5. data/README.md +265 -0
  6. data/Rakefile +12 -0
  7. data/lib/java/LibSVM.jar +0 -0
  8. data/lib/java/SMOTE.jar +0 -0
  9. data/lib/java/balancedRandomForest.jar +0 -0
  10. data/lib/java/diffutils-1.3.0.jar +0 -0
  11. data/lib/java/libsvm.jar +0 -0
  12. data/lib/java/oneClassClassifier.jar +0 -0
  13. data/lib/java/realAdaBoost.jar +0 -0
  14. data/lib/java/swc-engine-1.1.0-jar-with-dependencies.jar +0 -0
  15. data/lib/java/sweble-wikitext-extractor.jar +0 -0
  16. data/lib/weka/classifiers/functions/lib_svm.rb +15 -0
  17. data/lib/weka/classifiers/meta/one_class_classifier.rb +25 -0
  18. data/lib/weka/classifiers/meta/real_ada_boost.rb +17 -0
  19. data/lib/weka/classifiers/trees/balanced_random_forest.rb +18 -0
  20. data/lib/weka/filters/supervised/instance/smote.rb +22 -0
  21. data/lib/wikipedia.rb +51 -0
  22. data/lib/wikipedia/vandalism_detection.rb +30 -0
  23. data/lib/wikipedia/vandalism_detection/algorithms.rb +18 -0
  24. data/lib/wikipedia/vandalism_detection/algorithms/kullback_leibler_divergence.rb +69 -0
  25. data/lib/wikipedia/vandalism_detection/classifier.rb +186 -0
  26. data/lib/wikipedia/vandalism_detection/configuration.rb +321 -0
  27. data/lib/wikipedia/vandalism_detection/diff.rb +27 -0
  28. data/lib/wikipedia/vandalism_detection/edit.rb +75 -0
  29. data/lib/wikipedia/vandalism_detection/evaluator.rb +606 -0
  30. data/lib/wikipedia/vandalism_detection/exceptions.rb +40 -0
  31. data/lib/wikipedia/vandalism_detection/feature_calculator.rb +89 -0
  32. data/lib/wikipedia/vandalism_detection/features.rb +67 -0
  33. data/lib/wikipedia/vandalism_detection/features/all_wordlists_frequency.rb +23 -0
  34. data/lib/wikipedia/vandalism_detection/features/all_wordlists_impact.rb +22 -0
  35. data/lib/wikipedia/vandalism_detection/features/anonymity.rb +19 -0
  36. data/lib/wikipedia/vandalism_detection/features/anonymity_previous.rb +29 -0
  37. data/lib/wikipedia/vandalism_detection/features/article_size.rb +18 -0
  38. data/lib/wikipedia/vandalism_detection/features/bad_frequency.rb +23 -0
  39. data/lib/wikipedia/vandalism_detection/features/bad_impact.rb +21 -0
  40. data/lib/wikipedia/vandalism_detection/features/base.rb +54 -0
  41. data/lib/wikipedia/vandalism_detection/features/biased_frequency.rb +23 -0
  42. data/lib/wikipedia/vandalism_detection/features/biased_impact.rb +22 -0
  43. data/lib/wikipedia/vandalism_detection/features/blanking.rb +25 -0
  44. data/lib/wikipedia/vandalism_detection/features/character_diversity.rb +25 -0
  45. data/lib/wikipedia/vandalism_detection/features/character_sequence.rb +19 -0
  46. data/lib/wikipedia/vandalism_detection/features/comment_bad_frequency.rb +22 -0
  47. data/lib/wikipedia/vandalism_detection/features/comment_biased_frequency.rb +22 -0
  48. data/lib/wikipedia/vandalism_detection/features/comment_length.rb +17 -0
  49. data/lib/wikipedia/vandalism_detection/features/comment_markup_frequency.rb +27 -0
  50. data/lib/wikipedia/vandalism_detection/features/comment_pronoun_frequency.rb +22 -0
  51. data/lib/wikipedia/vandalism_detection/features/comment_sex_frequency.rb +22 -0
  52. data/lib/wikipedia/vandalism_detection/features/comment_vulgarism_frequency.rb +22 -0
  53. data/lib/wikipedia/vandalism_detection/features/compressibility.rb +27 -0
  54. data/lib/wikipedia/vandalism_detection/features/contains_base.rb +18 -0
  55. data/lib/wikipedia/vandalism_detection/features/copyedit.rb +18 -0
  56. data/lib/wikipedia/vandalism_detection/features/digit_ratio.rb +24 -0
  57. data/lib/wikipedia/vandalism_detection/features/edits_per_user.rb +65 -0
  58. data/lib/wikipedia/vandalism_detection/features/emoticons_frequency.rb +27 -0
  59. data/lib/wikipedia/vandalism_detection/features/emoticons_impact.rb +29 -0
  60. data/lib/wikipedia/vandalism_detection/features/frequency_base.rb +20 -0
  61. data/lib/wikipedia/vandalism_detection/features/impact_base.rb +22 -0
  62. data/lib/wikipedia/vandalism_detection/features/inserted_character_distribution.rb +22 -0
  63. data/lib/wikipedia/vandalism_detection/features/inserted_external_links.rb +18 -0
  64. data/lib/wikipedia/vandalism_detection/features/inserted_internal_links.rb +18 -0
  65. data/lib/wikipedia/vandalism_detection/features/inserted_size.rb +20 -0
  66. data/lib/wikipedia/vandalism_detection/features/inserted_words.rb +18 -0
  67. data/lib/wikipedia/vandalism_detection/features/longest_word.rb +20 -0
  68. data/lib/wikipedia/vandalism_detection/features/markup_frequency.rb +27 -0
  69. data/lib/wikipedia/vandalism_detection/features/markup_impact.rb +27 -0
  70. data/lib/wikipedia/vandalism_detection/features/non_alphanumeric_ratio.rb +24 -0
  71. data/lib/wikipedia/vandalism_detection/features/personal_life.rb +18 -0
  72. data/lib/wikipedia/vandalism_detection/features/pronoun_frequency.rb +23 -0
  73. data/lib/wikipedia/vandalism_detection/features/pronoun_impact.rb +21 -0
  74. data/lib/wikipedia/vandalism_detection/features/removed_all_wordlists_frequency.rb +23 -0
  75. data/lib/wikipedia/vandalism_detection/features/removed_bad_frequency.rb +23 -0
  76. data/lib/wikipedia/vandalism_detection/features/removed_biased_frequency.rb +23 -0
  77. data/lib/wikipedia/vandalism_detection/features/removed_character_distribution.rb +22 -0
  78. data/lib/wikipedia/vandalism_detection/features/removed_emoticons_frequency.rb +27 -0
  79. data/lib/wikipedia/vandalism_detection/features/removed_markup_frequency.rb +28 -0
  80. data/lib/wikipedia/vandalism_detection/features/removed_pronoun_frequency.rb +23 -0
  81. data/lib/wikipedia/vandalism_detection/features/removed_sex_frequency.rb +23 -0
  82. data/lib/wikipedia/vandalism_detection/features/removed_size.rb +20 -0
  83. data/lib/wikipedia/vandalism_detection/features/removed_vulgarism_frequency.rb +23 -0
  84. data/lib/wikipedia/vandalism_detection/features/removed_words.rb +18 -0
  85. data/lib/wikipedia/vandalism_detection/features/replacement_similarity.rb +22 -0
  86. data/lib/wikipedia/vandalism_detection/features/reverted.rb +18 -0
  87. data/lib/wikipedia/vandalism_detection/features/revisions_character_distribution.rb +21 -0
  88. data/lib/wikipedia/vandalism_detection/features/same_editor.rb +29 -0
  89. data/lib/wikipedia/vandalism_detection/features/sex_frequency.rb +23 -0
  90. data/lib/wikipedia/vandalism_detection/features/sex_impact.rb +21 -0
  91. data/lib/wikipedia/vandalism_detection/features/size_increment.rb +22 -0
  92. data/lib/wikipedia/vandalism_detection/features/size_ratio.rb +26 -0
  93. data/lib/wikipedia/vandalism_detection/features/term_frequency.rb +25 -0
  94. data/lib/wikipedia/vandalism_detection/features/time_interval.rb +31 -0
  95. data/lib/wikipedia/vandalism_detection/features/time_of_day.rb +22 -0
  96. data/lib/wikipedia/vandalism_detection/features/upper_case_ratio.rb +24 -0
  97. data/lib/wikipedia/vandalism_detection/features/upper_case_words_ratio.rb +31 -0
  98. data/lib/wikipedia/vandalism_detection/features/upper_to_lower_case_ratio.rb +24 -0
  99. data/lib/wikipedia/vandalism_detection/features/user_reputation.rb +38 -0
  100. data/lib/wikipedia/vandalism_detection/features/vulgarism_frequency.rb +23 -0
  101. data/lib/wikipedia/vandalism_detection/features/vulgarism_impact.rb +22 -0
  102. data/lib/wikipedia/vandalism_detection/features/weekday.rb +21 -0
  103. data/lib/wikipedia/vandalism_detection/features/words_increment.rb +22 -0
  104. data/lib/wikipedia/vandalism_detection/instances.rb +130 -0
  105. data/lib/wikipedia/vandalism_detection/page.rb +88 -0
  106. data/lib/wikipedia/vandalism_detection/page_parser.rb +52 -0
  107. data/lib/wikipedia/vandalism_detection/revision.rb +69 -0
  108. data/lib/wikipedia/vandalism_detection/revision_parser.rb +43 -0
  109. data/lib/wikipedia/vandalism_detection/test_dataset.rb +367 -0
  110. data/lib/wikipedia/vandalism_detection/text.rb +18 -0
  111. data/lib/wikipedia/vandalism_detection/training_dataset.rb +303 -0
  112. data/lib/wikipedia/vandalism_detection/version.rb +5 -0
  113. data/lib/wikipedia/vandalism_detection/wikitext_extractor.rb +80 -0
  114. data/lib/wikipedia/vandalism_detection/word_lists.rb +19 -0
  115. data/lib/wikipedia/vandalism_detection/word_lists/bad.rb +12 -0
  116. data/lib/wikipedia/vandalism_detection/word_lists/biased.rb +21 -0
  117. data/lib/wikipedia/vandalism_detection/word_lists/emoticons.rb +22 -0
  118. data/lib/wikipedia/vandalism_detection/word_lists/markup.rb +12 -0
  119. data/lib/wikipedia/vandalism_detection/word_lists/pronouns.rb +15 -0
  120. data/lib/wikipedia/vandalism_detection/word_lists/sex.rb +12 -0
  121. data/lib/wikipedia/vandalism_detection/word_lists/vulgarism.rb +97 -0
  122. data/spec/factories/edit.rb +20 -0
  123. data/spec/factories/page.rb +13 -0
  124. data/spec/factories/revision.rb +51 -0
  125. data/spec/resources/config/config.yml +35 -0
  126. data/spec/resources/corpora/test/edits.csv +8 -0
  127. data/spec/resources/corpora/test/ground-truth.txt +3 -0
  128. data/spec/resources/corpora/test/revisions/part-1/326471754.txt +199 -0
  129. data/spec/resources/corpora/test/revisions/part-1/326873205.txt +137 -0
  130. data/spec/resources/corpora/test/revisions/part-1/328774035.txt +162 -0
  131. data/spec/resources/corpora/test/revisions/part-2/307084144.txt +137 -0
  132. data/spec/resources/corpora/test/revisions/part-2/326978767.txt +199 -0
  133. data/spec/resources/corpora/test/revisions/part-2/328774110.txt +162 -0
  134. data/spec/resources/corpora/training/annotations.csv +7 -0
  135. data/spec/resources/corpora/training/edits.csv +7 -0
  136. data/spec/resources/corpora/training/revisions/part-1/326471754.txt +199 -0
  137. data/spec/resources/corpora/training/revisions/part-1/326873205.txt +137 -0
  138. data/spec/resources/corpora/training/revisions/part-1/328774035.txt +162 -0
  139. data/spec/resources/corpora/training/revisions/part-2/307084144.txt +137 -0
  140. data/spec/resources/corpora/training/revisions/part-2/326978767.txt +199 -0
  141. data/spec/resources/corpora/training/revisions/part-2/328774110.txt +162 -0
  142. data/spec/resources/page_with_redirects.xml +85 -0
  143. data/spec/resources/redirect_page.xml +59 -0
  144. data/spec/resources/revision_simplified.xml +13 -0
  145. data/spec/resources/sample_revision.txt +137 -0
  146. data/spec/resources/sample_revision_clean_text.txt +1 -0
  147. data/spec/resources/sample_revision_plain_text.txt +183 -0
  148. data/spec/resources/vandalism_on_wikipedia.xml +234 -0
  149. data/spec/resources/vandalism_on_wikipedia_simplified.xml +119 -0
  150. data/spec/resources/wikipedia_tokens.txt +30 -0
  151. data/spec/spec_helper.rb +22 -0
  152. data/spec/support/macros/file_reading.rb +7 -0
  153. data/spec/support/macros/test_configuration.rb +71 -0
  154. data/spec/vandalism_detection/algorithms/kullback_leibler_divergence_spec.rb +36 -0
  155. data/spec/vandalism_detection/classifier_spec.rb +317 -0
  156. data/spec/vandalism_detection/configuration_spec.rb +517 -0
  157. data/spec/vandalism_detection/diff_spec.rb +40 -0
  158. data/spec/vandalism_detection/edit_spec.rb +137 -0
  159. data/spec/vandalism_detection/evaluator_spec.rb +671 -0
  160. data/spec/vandalism_detection/feature_calculator_spec.rb +128 -0
  161. data/spec/vandalism_detection/features/all_wordlists_frequency_spec.rb +36 -0
  162. data/spec/vandalism_detection/features/all_wordlists_impact_spec.rb +58 -0
  163. data/spec/vandalism_detection/features/anonymity_previous_spec.rb +61 -0
  164. data/spec/vandalism_detection/features/anonymity_spec.rb +23 -0
  165. data/spec/vandalism_detection/features/article_size_spec.rb +35 -0
  166. data/spec/vandalism_detection/features/bad_frequency_spec.rb +36 -0
  167. data/spec/vandalism_detection/features/bad_impact_spec.rb +59 -0
  168. data/spec/vandalism_detection/features/base_spec.rb +49 -0
  169. data/spec/vandalism_detection/features/biased_frequency_spec.rb +36 -0
  170. data/spec/vandalism_detection/features/biased_impact_spec.rb +58 -0
  171. data/spec/vandalism_detection/features/blanking_spec.rb +38 -0
  172. data/spec/vandalism_detection/features/character_diversity_spec.rb +35 -0
  173. data/spec/vandalism_detection/features/character_sequence_spec.rb +37 -0
  174. data/spec/vandalism_detection/features/comment_bad_frequency_spec.rb +34 -0
  175. data/spec/vandalism_detection/features/comment_biased_frequency_spec.rb +34 -0
  176. data/spec/vandalism_detection/features/comment_length_spec.rb +27 -0
  177. data/spec/vandalism_detection/features/comment_markup_frequency_spec.rb +34 -0
  178. data/spec/vandalism_detection/features/comment_pronoun_frequency_spec.rb +34 -0
  179. data/spec/vandalism_detection/features/comment_sex_frequency_spec.rb +34 -0
  180. data/spec/vandalism_detection/features/comment_vulgarism_frequency_spec.rb +34 -0
  181. data/spec/vandalism_detection/features/compressibility_spec.rb +42 -0
  182. data/spec/vandalism_detection/features/contains_base_spec.rb +33 -0
  183. data/spec/vandalism_detection/features/copyedit_spec.rb +33 -0
  184. data/spec/vandalism_detection/features/digit_ratio_spec.rb +35 -0
  185. data/spec/vandalism_detection/features/edits_per_user_spec.rb +49 -0
  186. data/spec/vandalism_detection/features/emoticons_frequency_spec.rb +36 -0
  187. data/spec/vandalism_detection/features/emoticons_impact_spec.rb +51 -0
  188. data/spec/vandalism_detection/features/frequency_base_spec.rb +26 -0
  189. data/spec/vandalism_detection/features/impact_base_spec.rb +41 -0
  190. data/spec/vandalism_detection/features/inserted_character_distribution_spec.rb +46 -0
  191. data/spec/vandalism_detection/features/inserted_external_links_spec.rb +35 -0
  192. data/spec/vandalism_detection/features/inserted_internal_links_spec.rb +35 -0
  193. data/spec/vandalism_detection/features/inserted_size_spec.rb +35 -0
  194. data/spec/vandalism_detection/features/inserted_words_spec.rb +35 -0
  195. data/spec/vandalism_detection/features/longest_word_spec.rb +35 -0
  196. data/spec/vandalism_detection/features/markup_frequency_spec.rb +36 -0
  197. data/spec/vandalism_detection/features/markup_impact_spec.rb +59 -0
  198. data/spec/vandalism_detection/features/non_alphanumeric_ratio_spec.rb +35 -0
  199. data/spec/vandalism_detection/features/personal_life_spec.rb +26 -0
  200. data/spec/vandalism_detection/features/pronoun_frequency_spec.rb +36 -0
  201. data/spec/vandalism_detection/features/pronoun_impact_spec.rb +59 -0
  202. data/spec/vandalism_detection/features/removed_all_wordlists_frequency_spec.rb +36 -0
  203. data/spec/vandalism_detection/features/removed_bad_frequency_spec.rb +36 -0
  204. data/spec/vandalism_detection/features/removed_biased_frequency_spec.rb +36 -0
  205. data/spec/vandalism_detection/features/removed_character_distribution_spec.rb +46 -0
  206. data/spec/vandalism_detection/features/removed_emoticons_frequency_spec.rb +36 -0
  207. data/spec/vandalism_detection/features/removed_markup_frequency_spec.rb +36 -0
  208. data/spec/vandalism_detection/features/removed_pronoun_frequency_spec.rb +36 -0
  209. data/spec/vandalism_detection/features/removed_sex_frequency_spec.rb +36 -0
  210. data/spec/vandalism_detection/features/removed_size_spec.rb +35 -0
  211. data/spec/vandalism_detection/features/removed_vulgarism_frequency_spec.rb +36 -0
  212. data/spec/vandalism_detection/features/removed_words_spec.rb +35 -0
  213. data/spec/vandalism_detection/features/replacement_similarity_spec.rb +44 -0
  214. data/spec/vandalism_detection/features/reverted_spec.rb +28 -0
  215. data/spec/vandalism_detection/features/revisions_character_distribution_spec.rb +46 -0
  216. data/spec/vandalism_detection/features/same_editor_spec.rb +60 -0
  217. data/spec/vandalism_detection/features/sex_frequency_spec.rb +36 -0
  218. data/spec/vandalism_detection/features/sex_impact_spec.rb +59 -0
  219. data/spec/vandalism_detection/features/size_increment_spec.rb +35 -0
  220. data/spec/vandalism_detection/features/size_ratio_spec.rb +57 -0
  221. data/spec/vandalism_detection/features/term_frequency_spec.rb +38 -0
  222. data/spec/vandalism_detection/features/time_interval_spec.rb +50 -0
  223. data/spec/vandalism_detection/features/time_of_day_spec.rb +22 -0
  224. data/spec/vandalism_detection/features/upper_case_ratio_spec.rb +35 -0
  225. data/spec/vandalism_detection/features/upper_case_words_ratio_spec.rb +37 -0
  226. data/spec/vandalism_detection/features/upper_to_lower_case_ratio_spec.rb +35 -0
  227. data/spec/vandalism_detection/features/user_reputation_spec.rb +52 -0
  228. data/spec/vandalism_detection/features/vulgarism_frequency_spec.rb +36 -0
  229. data/spec/vandalism_detection/features/vulgarism_impact_spec.rb +58 -0
  230. data/spec/vandalism_detection/features/weekday_spec.rb +22 -0
  231. data/spec/vandalism_detection/features/words_increment_spec.rb +35 -0
  232. data/spec/vandalism_detection/instances_spec.rb +156 -0
  233. data/spec/vandalism_detection/page_parser_spec.rb +184 -0
  234. data/spec/vandalism_detection/page_spec.rb +135 -0
  235. data/spec/vandalism_detection/revision_parser_spec.rb +53 -0
  236. data/spec/vandalism_detection/revision_spec.rb +115 -0
  237. data/spec/vandalism_detection/test_dataset_spec.rb +231 -0
  238. data/spec/vandalism_detection/text_spec.rb +29 -0
  239. data/spec/vandalism_detection/training_dataset_spec.rb +264 -0
  240. data/spec/vandalism_detection/wikitext_extractor_spec.rb +72 -0
  241. data/spec/weka/classifiers/functions/lib_svm_spec.rb +38 -0
  242. data/spec/weka/classifiers/meta/one_class_classifier_spec.rb +76 -0
  243. data/spec/weka/classifiers/meta/real_ada_boost_spec.rb +40 -0
  244. data/spec/weka/classifiers/trees/balanced_random_forest_spec.rb +40 -0
  245. data/spec/weka/filters/supervised/instance/smote_spec.rb +6 -0
  246. data/wikipedia-vandalism_detection.gemspec +30 -0
  247. metadata +512 -0
@@ -0,0 +1,36 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::BiasedFrequency do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::BiasedFrequency.new
7
+ end
8
+
9
+ it { should be_a Wikipedia::VandalismDetection::Features::FrequencyBase }
10
+
11
+ describe "#calculate" do
12
+
13
+ it "returns the inserted number of biased words relative to all inserted words count" do
14
+ # inserted: total 10 words, 4 biased
15
+ old_text = Wikipedia::VandalismDetection::Text.new 'Great old.'
16
+ new_text = Wikipedia::VandalismDetection::Text.new "Great old. It's Great man, this is amazing, really a classic."
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 4.0 / 9.0
23
+ end
24
+
25
+ it "returns 0.0 on emtpy clean inserted text" do
26
+ old_text = Wikipedia::VandalismDetection::Text.new 'Great old.'
27
+ new_text = Wikipedia::VandalismDetection::Text.new "Great 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 +1,58 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::BiasedImpact do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::BiasedImpact.new
7
+ end
8
+
9
+ it { should be_a Wikipedia::VandalismDetection::Features::ImpactBase }
10
+
11
+ describe "#calculate" do
12
+
13
+ it "returns the impact of biased words of the edit's new revision text" do
14
+ # 1 vulgarism
15
+ old_text = Wikipedia::VandalismDetection::Text.new "this is classic!"
16
+ # 3 vulgarism
17
+ new_text = Wikipedia::VandalismDetection::Text.new "This is classic, legendary and amazing!"
18
+
19
+ old_revision = build(:old_revision, text: old_text)
20
+ new_revision = build(:new_revision, text: new_text)
21
+ edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
22
+
23
+ expect(@feature.calculate(edit)).to eq 1.0 / (1.0 + 3.0)
24
+ end
25
+
26
+ it "returns 0.5 on both no terms in text revisions" do
27
+ text = Wikipedia::VandalismDetection::Text.new "{{speedy deletion}}"
28
+
29
+ old_revision = build(:old_revision, text: text)
30
+ new_revision = build(:new_revision, text: text)
31
+ edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
32
+
33
+ expect(@feature.calculate(edit)).to eq 0.5
34
+ end
35
+
36
+ it "returns 0.0 on emtpy clean text of old revision" do
37
+ old_text = Wikipedia::VandalismDetection::Text.new "{{speedy deletion}}"
38
+ new_text = Wikipedia::VandalismDetection::Text.new "great"
39
+
40
+ old_revision = build(:old_revision, text: old_text)
41
+ new_revision = build(:new_revision, text: new_text)
42
+ edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
43
+
44
+ expect(@feature.calculate(edit)).to eq 0.0
45
+ end
46
+
47
+ it "returns 1.0 on emtpy clean text of new revision" do
48
+ old_text = Wikipedia::VandalismDetection::Text.new "great"
49
+ new_text = Wikipedia::VandalismDetection::Text.new "{{speedy deletion}}"
50
+
51
+ old_revision = build(:old_revision, text: old_text)
52
+ new_revision = build(:new_revision, text: new_text)
53
+ edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
54
+
55
+ expect(@feature.calculate(edit)).to eq 1.0
56
+ end
57
+ end
58
+ end
@@ -0,0 +1,38 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::Blanking do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::Blanking.new
7
+ @blank_text = "a" * (Wikipedia::VandalismDetection::Features::Blanking::BLANKING_THRESHOLD - 1)
8
+ end
9
+
10
+ it { should be_a Wikipedia::VandalismDetection::Features::Base }
11
+
12
+ describe "#calculate" do
13
+
14
+ it "returns 1.0 in case of full blanking (size < BLANKING_THRESHOLD) new revision" do
15
+ old_revision = build(:old_revision, text: "#{@blank_text} additional text")
16
+ new_revision = build(:new_revision, text: @blank_text)
17
+ edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
18
+
19
+ expect(@feature.calculate(edit)).to eq 1.0
20
+ end
21
+
22
+ it "returns 0.0 in case of not full blanking (size >= BLANKING_THRESHOLD) new revision" do
23
+ old_revision = build(:old_revision, text: "#{@blank_text} additional text")
24
+ new_revision = build(:new_revision, text: "#{@blank_text}a")
25
+ edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
26
+
27
+ expect(@feature.calculate(edit)).to eq 0.0
28
+ end
29
+
30
+ it "returns 0.0 if old revision is <= new revision" do
31
+ old_revision = build(:old_revision, text: @blank_text)
32
+ new_revision = build(:new_revision, text: @blank_text.next!)
33
+ edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
34
+
35
+ expect(@feature.calculate(edit)).to eq 0.0
36
+ end
37
+ end
38
+ end
@@ -0,0 +1,35 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::CharacterDiversity do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::CharacterDiversity.new
7
+ end
8
+
9
+ it { should be_a Wikipedia::VandalismDetection::Features::Base }
10
+
11
+ describe "#calculate" do
12
+
13
+ it "returns the character diversity of the edit's new revision inserted text" do
14
+ old_text = Wikipedia::VandalismDetection::Text.new('text')
15
+ new_text = Wikipedia::VandalismDetection::Text.new 'text [[aa ab cdeefg]]' # 9 unique characters of total 14
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 14 ** (1.0 / 9)
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 +1,37 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::CharacterSequence do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::CharacterSequence.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 the new revision's longest sequence of the same character" do
14
+ old_text = Wikipedia::VandalismDetection::Text.new "a 666666"
15
+ new_text = Wikipedia::VandalismDetection::Text.new "a 666666 4444ccc eeeefffff gggg g \n===hhhh===\n''fffaffff''"
16
+
17
+ old_revision = build(:old_revision, text: old_text)
18
+ new_revision = build(:new_revision, text: new_text)
19
+
20
+ edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
21
+
22
+ expect(@feature.calculate(edit)).to eq 5
23
+ end
24
+
25
+ it "returns 0 on non inserted text" do
26
+ old_text = Wikipedia::VandalismDetection::Text.new "a 666666 4444ccc eeeefffff gggg g"
27
+ new_text = Wikipedia::VandalismDetection::Text.new "a 666666 "
28
+
29
+ old_revision = build(:old_revision, text: old_text)
30
+ new_revision = build(:new_revision, text: new_text)
31
+
32
+ edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
33
+
34
+ expect(@feature.calculate(edit)).to eq 0
35
+ end
36
+ end
37
+ end
@@ -0,0 +1,34 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::CommentBadFrequency do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::CommentBadFrequency.new
7
+ end
8
+
9
+ it { should be_a Wikipedia::VandalismDetection::Features::FrequencyBase }
10
+
11
+ describe "#calculate" do
12
+
13
+ it "returns the number of bad words in comment relative to all words count" do
14
+ # total 11 words, 7 bad words
15
+ comment = Wikipedia::VandalismDetection::Text.new '666 was 666 if 666 was 666 and guy are 666'
16
+
17
+ old_revision = build(:old_revision)
18
+ new_revision = build(:new_revision, comment: comment)
19
+ edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
20
+
21
+ expect(@feature.calculate(edit)).to eq 6.0 / 11.0
22
+ end
23
+
24
+ it "returns 0.0 on emtpy clean text comment" do
25
+ comment = Wikipedia::VandalismDetection::Text.new "{{speedy deletion}}"
26
+
27
+ old_revision = build(:old_revision)
28
+ new_revision = build(:new_revision, comment: comment)
29
+ edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
30
+
31
+ expect(@feature.calculate(edit)).to eq 0.0
32
+ end
33
+ end
34
+ end
@@ -0,0 +1,34 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::CommentBiasedFrequency do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::CommentBiasedFrequency.new
7
+ end
8
+
9
+ it { should be_a Wikipedia::VandalismDetection::Features::FrequencyBase }
10
+
11
+ describe "#calculate" do
12
+
13
+ it "returns the number of biased words in comment relative to all words count" do
14
+ # total 10 words, 3 biased
15
+ comment = Wikipedia::VandalismDetection::Text.new "It's Great man, this is amazing, really a classic."
16
+
17
+ old_revision = build(:old_revision)
18
+ new_revision = build(:new_revision, comment: comment)
19
+ edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
20
+
21
+ expect(@feature.calculate(edit)).to eq 4.0 / 9.0
22
+ end
23
+
24
+ it "returns 0.0 on emtpy clean text comment" do
25
+ comment = Wikipedia::VandalismDetection::Text.new "{{speedy deletion}}"
26
+
27
+ old_revision = build(:old_revision)
28
+ new_revision = build(:new_revision, comment: comment)
29
+ edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
30
+
31
+ expect(@feature.calculate(edit)).to eq 0.0
32
+ end
33
+ end
34
+ end
@@ -0,0 +1,27 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::CommentLength do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::CommentLength.new
7
+ end
8
+
9
+ it { should be_a Wikipedia::VandalismDetection::Features::Base }
10
+
11
+ describe "#calculate" do
12
+
13
+ it "returns the length of the edit's new revisions comment" do
14
+ comment = Wikipedia::VandalismDetection::Text.new "1 34567 9"
15
+ edit = build :edit, new_revision: build(:new_revision, comment: comment)
16
+
17
+ expect(@feature.calculate(edit)).to eq 9
18
+ end
19
+
20
+ it "returns 0 on emtpy clean text" do
21
+ text = Wikipedia::VandalismDetection::Text.new "{{speedy deletion}}"
22
+ edit = build :edit, new_revision: build(:new_revision, text: text)
23
+
24
+ expect(@feature.calculate(edit)).to eq 0
25
+ end
26
+ end
27
+ end
@@ -0,0 +1,34 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::CommentMarkupFrequency do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::CommentMarkupFrequency.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 markup words in comment relative to all words count" do
14
+ # total 7 words, 3 markup
15
+ comment = Wikipedia::VandalismDetection::Text.new '[[Content]] is not always {{simple}} to [[produce]]'
16
+
17
+ old_revision = build(:old_revision)
18
+ new_revision = build(:new_revision, comment: comment)
19
+ edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
20
+
21
+ expect(@feature.calculate(edit)).to eq 3.0 / 7.0
22
+ end
23
+
24
+ it "returns 0.0 on emtpy text comment" do
25
+ comment = Wikipedia::VandalismDetection::Text.new ""
26
+
27
+ old_revision = build(:old_revision)
28
+ new_revision = build(:new_revision, comment: comment)
29
+ edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
30
+
31
+ expect(@feature.calculate(edit)).to eq 0.0
32
+ end
33
+ end
34
+ end
@@ -0,0 +1,34 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::CommentPronounFrequency do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::CommentPronounFrequency.new
7
+ end
8
+
9
+ it { should be_a Wikipedia::VandalismDetection::Features::FrequencyBase }
10
+
11
+ describe "#calculate" do
12
+
13
+ it "returns the number of pronouns in comment relative to all words count" do
14
+ # total 12 words, 7 pronouns
15
+ comment = Wikipedia::VandalismDetection::Text.new 'I was you if You was Me and we are ourselves us'
16
+
17
+ old_revision = build(:old_revision)
18
+ new_revision = build(:new_revision, comment: comment)
19
+ edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
20
+
21
+ expect(@feature.calculate(edit)).to eq 7.0 / 12.0
22
+ end
23
+
24
+ it "returns 0.0 on emtpy clean text comment" do
25
+ comment = Wikipedia::VandalismDetection::Text.new "{{speedy deletion}}"
26
+
27
+ old_revision = build(:old_revision)
28
+ new_revision = build(:new_revision, comment: comment)
29
+ edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
30
+
31
+ expect(@feature.calculate(edit)).to eq 0.0
32
+ end
33
+ end
34
+ end
@@ -0,0 +1,34 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::CommentSexFrequency do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::CommentSexFrequency.new
7
+ end
8
+
9
+ it { should be_a Wikipedia::VandalismDetection::Features::FrequencyBase }
10
+
11
+ describe "#calculate" do
12
+
13
+ it "returns the number of sex words in comment relative to all words count" do
14
+ # total 11 words, 7 sex words
15
+ comment = Wikipedia::VandalismDetection::Text.new 'Penis was penis if penis was penis and penis are penis'
16
+
17
+ old_revision = build(:old_revision)
18
+ new_revision = build(:new_revision, comment: comment)
19
+ edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
20
+
21
+ expect(@feature.calculate(edit)).to eq 6.0 / 11.0
22
+ end
23
+
24
+ it "returns 0.0 on emtpy clean text comment" do
25
+ comment = Wikipedia::VandalismDetection::Text.new "{{speedy deletion}}"
26
+
27
+ old_revision = build(:old_revision)
28
+ new_revision = build(:new_revision, comment: comment)
29
+ edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
30
+
31
+ expect(@feature.calculate(edit)).to eq 0.0
32
+ end
33
+ end
34
+ end
@@ -0,0 +1,34 @@
1
+ require 'spec_helper'
2
+
3
+ describe Wikipedia::VandalismDetection::Features::CommentVulgarismFrequency do
4
+
5
+ before do
6
+ @feature = Wikipedia::VandalismDetection::Features::CommentVulgarismFrequency.new
7
+ end
8
+
9
+ it { should be_a Wikipedia::VandalismDetection::Features::FrequencyBase }
10
+
11
+ describe "#calculate" do
12
+
13
+ it "returns the number of vulgarism words in comment relative to all words count" do
14
+ # total 7 words, 2 vulgarism
15
+ comment = Wikipedia::VandalismDetection::Text.new "Fuck you bitch. This is my change!"
16
+
17
+ old_revision = build(:old_revision)
18
+ new_revision = build(:new_revision, comment: comment)
19
+ edit = build(:edit, old_revision: old_revision, new_revision: new_revision)
20
+
21
+ expect(@feature.calculate(edit)).to eq 2.0 / 7.0
22
+ end
23
+
24
+ it "returns 0.0 on emtpy clean text comment" do
25
+ comment = Wikipedia::VandalismDetection::Text.new "{{speedy deletion}}"
26
+
27
+ old_revision = build(:old_revision)
28
+ new_revision = build(:new_revision, comment: comment)
29
+ edit = build(:edit, new_revision: new_revision, old_revision: old_revision)
30
+
31
+ expect(@feature.calculate(edit)).to eq 0.0
32
+ end
33
+ end
34
+ end