def test_log():
log_five = modifiers.log(5)
assert solve(log_five) == math_log(5)
assert solve(pickle.loads(pickle.dumps(log_five))) == math_log(5)
assert repr(log_five) == "<feature.log(5)>"
def test_trim():
d1 = Datasource("derp1")
f1 = Feature("foobar1", returns=int)
f2 = Feature("foobar2", returns=int, depends_on=[d1])
c = Constant(value=5)
fv = FeatureVector("foobar3", returns=int, depends_on=[c])
assert list(trim(f1)) == [f1]
assert list(trim([f1, f2, fv])) == [f1, f2, fv]
assert (list(trim(log(max(f1 - f2, 1)))) == [f1, f2])
def test_trim():
d1 = Datasource("derp1")
f1 = Feature("foobar1", returns=int)
f2 = Feature("foobar2", returns=int, depends_on=[d1])
c = Constant(value=5)
fv = FeatureVector("foobar3", returns=int, depends_on=[c])
assert list(trim(f1)) == [f1]
assert list(trim([f1, f2, fv])) == [f1, f2, fv]
assert (list(trim(log(max(f1 - f2, 1)))) ==
[f1, f2])
from revscoring.features import Feature, wikitext
from revscoring.features.modifiers import div, log, max, sub
def _process_new_longest(p_longest, r_longest):
if r_longest > p_longest:
return r_longest
else:
return 1
parent = [
log(wikitext.revision.parent.chars + 1),
log(wikitext.revision.parent.tokens + 1),
log(wikitext.revision.parent.words + 1),
log(wikitext.revision.parent.uppercase_words + 1),
log(wikitext.revision.parent.headings + 1),
log(wikitext.revision.parent.wikilinks + 1),
log(wikitext.revision.parent.external_links + 1),
log(wikitext.revision.parent.templates + 1),
log(wikitext.revision.parent.ref_tags + 1),
div(wikitext.revision.parent.chars,
max(wikitext.revision.parent.words, 1),
name="revision.parent.chars_per_word"),
div(wikitext.revision.parent.words,
max(wikitext.revision.parent.tokens, 1),
name="revision.parent.words_per_token"),
div(wikitext.revision.parent.uppercase_words,
max(wikitext.revision.parent.words, 1),
name="revision.parent.uppercase_words_per_word"),
div(wikitext.revision.parent.markups,
from revscoring.features import (
diff, page, parent_revision, previous_user_revision, revision, user
)
from revscoring.features.modifiers import log
from . import generic
damaging = generic.damaging + [
log(max(diff.added_badwords_ratio + 1,1)),
log(max(diff.added_misspellings_ratio + 1,1)),
log(max(diff.badwords_added + 1,1)),
log(max(diff.badwords_removed + 1,1)),
log(max(diff.misspellings_added + 1,1)),
log(max(diff.misspellings_removed + 1,1)),
log(max(diff.proportion_of_badwords_added + 1,1)),
log(max(diff.proportion_of_badwords_removed + 1,1)),
log(max(diff.proportion_of_misspellings_added + 1,1)),
log(max(diff.proportion_of_misspellings_removed + 1,1)),
log(max(diff.removed_badwords_ratio + 1,1)),
log(max(diff.removed_misspellings_ratio + 1,1)),
log(max(parent_revision.badwords + 1,1)),
log(max(parent_revision.misspellings + 1,1)),
log(max(parent_revision.proportion_of_badwords + 1,1)),
log(max(parent_revision.proportion_of_misspellings + 1,1)),
log(max(revision.badwords + 1,1)),
log(max(revision.misspellings + 1,1)),
log(max(revision.proportion_of_badwords + 1,1)),
log(max(revision.proportion_of_misspellings + 1,1)),
log(revision.infonoise + 1)
]
from revscoring.features import diff, page, parent_revision, user
from revscoring.features.modifiers import log
no_lang_damaging = [
log(diff.added_symbolic_chars_ratio + 1),
log(diff.chars_added + 1),
log(diff.chars_removed + 1),
diff.longest_repeated_char_added,
diff.longest_token_added,
log(diff.markup_chars_added + 1),
log(diff.markup_chars_removed + 1),
log(diff.numeric_chars_added + 1),
log(diff.numeric_chars_removed + 1),
diff.proportion_of_chars_added,
diff.proportion_of_chars_removed,
diff.proportion_of_markup_chars_added,
diff.proportion_of_numeric_chars_added,
diff.proportion_of_symbolic_chars_added,
diff.proportion_of_uppercase_chars_added,
log(diff.segments_added + 1),
log(diff.segments_removed + 1),
log(diff.symbolic_chars_added + 1),
log(diff.symbolic_chars_removed + 1),
log(diff.uppercase_chars_added + 1),
log(diff.uppercase_chars_removed + 1),
diff.bytes_changed + 1,
diff.bytes_changed_ratio,
page.is_content_namespace,
parent_revision.was_same_user,
user.is_bot
]
category_links / max(wikitext.revision.content_chars, 1),
cite_templates,
cite_templates / max(wikitext.revision.content_chars, 1),
proportion_of_templated_references,
non_templated_references,
non_templated_references / max(wikitext.revision.content_chars, 1),
non_cite_templates,
non_cite_templates / max(wikitext.revision.content_chars, 1),
infobox_templates,
cn_templates + 1,
cn_templates / max(wikitext.revision.content_chars, 1),
who_templates + 1,
who_templates / max(wikitext.revision.content_chars, 1),
main_article_templates,
main_article_templates / max(wikitext.revision.content_chars, 1),
(english.stemmed.revision.stem_chars /
max(wikitext.revision.content_chars, 1)),
log(paragraphs_without_refs_total_length + 1),
]
wp10 = wikipedia.article + local_wiki
"""
Based largely on work by Morten Warncke-Wang et al.[1] and with a few
improvements and extensions that Morten identified after publication.
1. Warncke-Wang, M., Cosley, D., & Riedl, J. (2013, August). Tell me more: An
actionable quality model for wikipedia. In Proceedings of the 9th
International Symposium on Open Collaboration (p. 8). ACM.
http://opensym.org/wsos2013/proceedings/p0202-warncke.pdf
"""
max(french.diff.words_removed, 1)
proportion_of_misspellings_added = french.diff.misspellings_added / \
max(french.diff.words_added, 1)
proportion_of_misspellings_removed = french.diff.misspellings_removed / \
max(french.diff.words_removed, 1)
proportion_of_badwords = french.parent_revision.badwords / \
max(french.parent_revision.words, 1)
proportion_of_misspellings = french.parent_revision.misspellings / \
max(french.parent_revision.words, 1)
added_badwords_ratio = proportion_of_badwords_added / \
max(proportion_of_badwords, 0.01)
added_misspellings_ratio = proportion_of_misspellings_added / \
max(proportion_of_misspellings, 0.01)
damaging = util.no_lang_damaging + enwiki.badwords + enwiki.informals + [
log(french.diff.badwords_added + 1),
log(french.diff.badwords_removed + 1),
log(french.diff.misspellings_added + 1),
log(french.diff.misspellings_removed + 1),
proportion_of_badwords_added,
proportion_of_badwords_removed,
proportion_of_misspellings_added,
proportion_of_misspellings_removed,
added_badwords_ratio,
added_misspellings_ratio
]
goodfaith = damaging
from revscoring.features import wikibase
from revscoring.features.modifiers import log
item = [
log(wikibase.revision.claims + 1),
log(wikibase.revision.properties + 1),
log(wikibase.revision.aliases + 1),
log(wikibase.revision.sources + 1),
log(wikibase.revision.qualifiers + 1),
log(wikibase.revision.badges + 1),
log(wikibase.revision.labels + 1),
log(wikibase.revision.sitelinks + 1),
log(wikibase.revision.descriptions + 1)
]
proportion_of_informals_removed = estonian.diff.informals_removed / max(estonian.diff.words_removed, 1)
proportion_of_badwords = estonian.parent_revision.badwords / max(estonian.parent_revision.words, 1)
proportion_of_misspellings = estonian.parent_revision.misspellings / max(estonian.parent_revision.words, 1)
proportion_of_informals = estonian.parent_revision.informals / max(estonian.parent_revision.words, 1)
added_badwords_ratio = proportion_of_badwords_added / max(proportion_of_badwords, 0.01)
added_misspellings_ratio = proportion_of_misspellings_added / max(proportion_of_misspellings, 0.01)
added_informals_ratio = proportion_of_informals_added / max(proportion_of_informals, 0.01)
damaging = (
util.no_lang_damaging
+ enwiki.badwords
+ enwiki.informals
+ [
log(estonian.diff.badwords_added + 1),
log(estonian.diff.badwords_removed + 1),
log(estonian.diff.informals_added + 1),
log(estonian.diff.informals_removed + 1),
log(estonian.diff.misspellings_added + 1),
log(estonian.diff.misspellings_removed + 1),
proportion_of_badwords_added,
proportion_of_badwords_removed,
proportion_of_informals_added,
proportion_of_informals_removed,
proportion_of_misspellings_added,
proportion_of_misspellings_removed,
added_badwords_ratio,
added_informals_ratio,
added_misspellings_ratio,
]
# log(revscoring.features.diff.numeric_chars_added + 1),
# log(revscoring.features.diff.numeric_chars_removed + 1),
# revscoring.features.diff.proportion_of_chars_added,
# revscoring.features.diff.proportion_of_chars_removed,
# revscoring.features.diff.proportion_of_numeric_chars_added,
# revscoring.features.diff.proportion_of_symbolic_chars_added,
# revscoring.features.diff.proportion_of_uppercase_chars_added,
# log(revscoring.features.diff.symbolic_chars_added + 1),
# log(revscoring.features.diff.symbolic_chars_removed + 1),
# log(revscoring.features.diff.uppercase_chars_added + 1),
# log(revscoring.features.diff.uppercase_chars_removed + 1),
# revscoring.features.diff.bytes_changed + 1,
# revscoring.featuresdiff.bytes_changed_ratio,
# page.is_content_namespace,
# parent_revision.was_same_user,
log(user.age + 1),
diff.number_added_sitelinks,
diff.number_removed_sitelinks,
diff.number_changed_sitelinks,
diff.number_added_labels,
diff.number_removed_labels,
diff.number_changed_labels,
diff.number_added_descriptions,
diff.number_removed_descriptions,
diff.number_changed_descriptions,
diff.number_added_aliases,
diff.number_removed_aliases,
diff.number_added_claims,
diff.number_removed_claims,
diff.number_changed_claims,
diff.number_changed_identifiers,
revision_oriented.revision.user.in_group({'bot'},
name="revision.user.is_bot"),
revision_oriented.revision.user.in_group(
{'checkuser', 'bureaucrat', 'oversight', 'steward'},
name="revision.user.has_advanced_rights"),
revision_oriented.revision.user.in_group({'sysop'},
name="revision.user.is_admin"),
revision_oriented.revision.user.in_group(
{
'browsearchive', 'deletedhistory', 'interface-editor',
'noratelimit', 'accountcreator', 'massmessage-sender',
'templateeditor', 'autopatrolled', 'propertycreator',
'centralnoticeadmin'
},
name="revision.user.is_trusted"),
revision_oriented.revision.user.in_group(
{
'rollback', 'abusefilter', 'patroller', 'reviewer', 'autoreview',
'autoreviewer', 'editor', 'autoeditor', 'eliminator'
},
name="revision.user.is_patroller"),
revision_oriented.revision.user.in_group(
{'import', 'filemover', 'suppressredirect'},
name="revision.user.is_curator")
]
protected_user = [
revision_oriented.revision.user.is_anon,
log(temporal.revision.user.seconds_since_registration + 1)
]
]
user_rights = [
revision_oriented.revision.user.in_group(
{'bot'}, name="revision.user.is_bot"),
revision_oriented.revision.user.in_group(
{'checkuser', 'bureaucrat', 'oversight', 'steward'},
name="revision.user.has_advanced_rights"),
revision_oriented.revision.user.in_group(
{'sysop'}, name="revision.user.is_admin"),
revision_oriented.revision.user.in_group(
{'browsearchive', 'deletedhistory', 'interface-editor',
'noratelimit', 'accountcreator', 'massmessage-sender', 'templateeditor',
'autopatrolled', 'propertycreator', 'centralnoticeadmin'},
name="revision.user.is_trusted"),
revision_oriented.revision.user.in_group(
{'rollback', 'abusefilter', 'patroller',
'reviewer', 'autoreview', 'autoreviewer', 'editor', 'autoeditor',
'eliminator'},
name="revision.user.is_patroller"),
revision_oriented.revision.user.in_group(
{'import', 'filemover', 'suppressredirect'},
name="revision.user.is_curator")
]
protected_user = [
revision_oriented.revision.user.is_anon,
log(temporal.revision.user.seconds_since_registration + 1)
]
proportion_of_badwords = hebrew.parent_revision.badwords / \
max(hebrew.parent_revision.words, 1)
proportion_of_misspellings = hebrew.parent_revision.misspellings / \
max(hebrew.parent_revision.words, 1)
proportion_of_informals = hebrew.parent_revision.informals / \
max(hebrew.parent_revision.words, 1)
added_badwords_ratio = proportion_of_badwords_added / \
max(proportion_of_badwords, 0.01)
added_misspellings_ratio = proportion_of_misspellings_added / \
max(proportion_of_misspellings, 0.01)
added_informals_ratio = proportion_of_informals_added / \
max(proportion_of_informals, 0.01)
damaging = util.no_lang_damaging + enwiki.badwords + enwiki.informals + [
log(hebrew.diff.badwords_added + 1),
log(hebrew.diff.badwords_removed + 1),
log(hebrew.diff.informals_added + 1),
log(hebrew.diff.informals_removed + 1),
log(hebrew.diff.misspellings_added + 1),
log(hebrew.diff.misspellings_removed + 1),
proportion_of_badwords_added,
proportion_of_badwords_removed,
proportion_of_informals_added,
proportion_of_informals_removed,
proportion_of_misspellings_added,
proportion_of_misspellings_removed,
added_badwords_ratio,
added_informals_ratio,
added_misspellings_ratio
]
from revscoring.features import Feature, wikitext
from revscoring.features.modifiers import div, log, max, sub
def _process_new_longest(p_longest, r_longest):
if r_longest > p_longest:
return r_longest
else:
return 1
parent = [
log(wikitext.revision.parent.chars + 1),
log(wikitext.revision.parent.tokens + 1),
log(wikitext.revision.parent.words + 1),
log(wikitext.revision.parent.uppercase_words + 1),
log(wikitext.revision.parent.headings + 1),
log(wikitext.revision.parent.wikilinks + 1),
log(wikitext.revision.parent.external_links + 1),
log(wikitext.revision.parent.templates + 1),
log(wikitext.revision.parent.ref_tags + 1),
div(wikitext.revision.parent.chars,
max(wikitext.revision.parent.words, 1),
name="revision.parent.chars_per_word"),
div(wikitext.revision.parent.words,
max(wikitext.revision.parent.tokens, 1),
name="revision.parent.words_per_token"),
div(wikitext.revision.parent.uppercase_words,
max(wikitext.revision.parent.words, 1),
name="revision.parent.uppercase_words_per_word"),
div(wikitext.revision.parent.markups,
max(wikitext.revision.parent.tokens, 1),
"glwiki.revision.revision",
wikitext.revision.datasources,
)
paragraphs = mappers.map(
str, revision.paragraphs_sentences_and_whitespace,
name="glwiki.revision.paragraphs"
)
paragraphs_without_refs = filters.regex_matching(
r"^(?!\s*$)((?!<ref>)(.|\n))*$",
paragraphs,
name="glwiki.revision.paragraphs_without_refs"
)
paragraphs_without_refs_total_length = aggregators.sum(
mappers.map(len, paragraphs_without_refs),
name="glwiki.revision.paragraphs_without_refs_total_length"
)
local_wiki = [
image_links,
image_links / max(wikitext.revision.content_chars, 1),
category_links,
category_links / max(wikitext.revision.content_chars, 1),
cn_templates + 1,
cn_templates / max(wikitext.revision.content_chars, 1),
log(paragraphs_without_refs_total_length + 1),
paragraphs_without_refs_total_length /
max(wikitext.revision.content_chars, 1),
]
wp10 = wikipedia.article + local_wiki
proportion_of_badwords = indonesian.parent_revision.badwords / \
max(indonesian.parent_revision.words, 1)
proportion_of_misspellings = indonesian.parent_revision.misspellings / \
max(indonesian.parent_revision.words, 1)
proportion_of_informals = indonesian.parent_revision.informals / \
max(indonesian.parent_revision.words, 1)
added_badwords_ratio = proportion_of_badwords_added / \
max(proportion_of_badwords, 0.01)
added_misspellings_ratio = proportion_of_misspellings_added / \
max(proportion_of_misspellings, 0.01)
added_informals_ratio = proportion_of_informals_added / \
max(proportion_of_informals, 0.01)
damaging = util.no_lang_damaging + enwiki.badwords + enwiki.informals + [
log(indonesian.diff.badwords_added + 1),
log(indonesian.diff.badwords_removed + 1),
log(indonesian.diff.informals_added + 1),
log(indonesian.diff.informals_removed + 1),
log(indonesian.diff.misspellings_added + 1),
log(indonesian.diff.misspellings_removed + 1),
proportion_of_badwords_added,
proportion_of_badwords_removed,
proportion_of_informals_added,
proportion_of_informals_removed,
proportion_of_misspellings_added,
proportion_of_misspellings_removed,
added_badwords_ratio,
added_informals_ratio,
added_misspellings_ratio
]
r"Cn",
r"Fact"
]
cn_templates = templates_that_match("|".join(CN_TEMPLATES),
name="enwiki.revision.cn_templates")
who_templates = templates_that_match("Who",
name="enwiki.revision.cn_templates")
main_article_templates = templates_that_match(
"Main",
name="enwiki.main_article_templates")
wp10 = [
revision.category_links,
log(revision.content_chars + 1),
log(revision.image_links + 1),
revision.image_links / max(revision.content_chars, 1),
log(cite_templates + 1),
log((revision.templates - cite_templates) + 1),
infobox_templates,
english.revision.infonoise,
log(revision.internal_links + 1),
revision.internal_links / max(revision.content_chars, 1),
revision.level_2_headings,
revision.level_2_headings / max(revision.content_chars, 1),
revision.level_3_headings,
revision.level_3_headings / max(revision.content_chars, 1),
log(revision.ref_tags + 1),
revision.ref_tags / max(revision.content_chars, 1),
log(max((revision.ref_tags - cite_templates) + 1, 1)),
max(english.parent_revision.words, 1)
proportion_of_misspellings = english.parent_revision.misspellings / \
max(english.parent_revision.words, 1)
proportion_of_informals = english.parent_revision.informals / \
max(english.parent_revision.words, 1)
added_badwords_ratio = proportion_of_badwords_added / \
max(proportion_of_badwords, 0.01)
added_misspellings_ratio = proportion_of_misspellings_added / \
max(proportion_of_misspellings, 0.01)
added_informals_ratio = proportion_of_informals_added / \
max(proportion_of_informals, 0.01)
badwords = [
added_badwords_ratio,
log(english.diff.badwords_added + 1),
log(english.diff.badwords_removed + 1),
proportion_of_badwords_added,
proportion_of_badwords_removed,
]
informals = [
added_informals_ratio,
log(english.diff.informals_added + 1),
log(english.diff.informals_removed + 1),
proportion_of_informals_added,
proportion_of_informals_removed
]
damaging = util.no_lang_damaging + [
log(english.diff.words_added + 1),
max(turkish.diff.words_removed, 1)
proportion_of_informals_added = turkish.diff.informals_added / \
max(turkish.diff.words_added, 1)
proportion_of_informals_removed = turkish.diff.informals_removed / \
max(turkish.diff.words_removed, 1)
proportion_of_badwords = turkish.parent_revision.badwords / \
max(turkish.parent_revision.words, 1)
proportion_of_informals = turkish.parent_revision.informals / \
max(turkish.parent_revision.words, 1)
added_badwords_ratio = proportion_of_badwords_added / \
max(proportion_of_badwords, 0.01)
added_informals_ratio = proportion_of_informals_added / \
max(proportion_of_informals, 0.01)
damaging = util.no_lang_damaging + enwiki.badwords + enwiki.informals + [
log(turkish.diff.badwords_added + 1),
log(turkish.diff.badwords_removed + 1),
log(turkish.diff.informals_added + 1),
log(turkish.diff.informals_removed + 1),
proportion_of_badwords_added,
proportion_of_badwords_removed,
proportion_of_informals_added,
proportion_of_informals_removed,
added_badwords_ratio,
added_informals_ratio
]
goodfaith = damaging
proportion_of_badwords = italian.parent_revision.badwords / \
max(italian.parent_revision.words, 1)
proportion_of_misspellings = italian.parent_revision.misspellings / \
max(italian.parent_revision.words, 1)
proportion_of_informals = italian.parent_revision.informals / \
max(italian.parent_revision.words, 1)
added_badwords_ratio = proportion_of_badwords_added / \
max(proportion_of_badwords, 0.01)
added_misspellings_ratio = proportion_of_misspellings_added / \
max(proportion_of_misspellings, 0.01)
added_informals_ratio = proportion_of_informals_added / \
max(proportion_of_informals, 0.01)
damaging = util.no_lang_damaging + english.badwords + [
log(italian.diff.words_added + 1),
log(italian.diff.words_removed + 1),
log(italian.parent_revision.words + 1),
log(italian.diff.badwords_added + 1),
log(italian.diff.badwords_removed + 1),
log(italian.diff.informals_added + 1),
log(italian.diff.informals_removed + 1),
log(italian.diff.misspellings_added + 1),
log(italian.diff.misspellings_removed + 1),
proportion_of_badwords_added,
proportion_of_badwords_removed,
proportion_of_informals_added,
proportion_of_informals_removed,
proportion_of_misspellings_added,
proportion_of_misspellings_removed,
added_badwords_ratio,
from revscoring.features import (
diff, page, parent_revision, previous_user_revision, revision, user
)
from revscoring.features.modifiers import log
from . import generic
damaging = generic.damaging + [
log(diff.added_badwords_ratio + 1),
log(diff.badwords_added + 1),
log(diff.badwords_removed + 1),
log(diff.proportion_of_badwords_added + 1),
log(diff.proportion_of_badwords_removed + 1),
log(diff.removed_badwords_ratio + 1),
log(parent_revision.badwords + 1),
log(parent_revision.proportion_of_badwords + 1),
log(revision.badwords + 1),
log(revision.proportion_of_badwords + 1)
]
good_faith = generic.good_faith + [
log(diff.added_badwords_ratio + 1),
log(diff.badwords_added + 1),
log(diff.badwords_removed + 1),
log(diff.proportion_of_badwords_added + 1),
log(diff.proportion_of_badwords_removed + 1),
log(diff.removed_badwords_ratio + 1),
log(parent_revision.badwords + 1),
log(parent_revision.proportion_of_badwords + 1),
log(revision.badwords + 1),
log(revision.proportion_of_badwords + 1)
]
all_images, all_images / max(wikitext.revision.content_chars, 1),
category_links, category_links / max(wikitext.revision.content_chars, 1),
all_ref_tags, all_ref_tags / max(wikitext.revision.content_chars, 1),
all_cite_templates,
all_cite_templates / max(wikitext.revision.content_chars, 1),
proportion_of_templated_references, non_templated_references,
non_templated_references / max(wikitext.revision.content_chars, 1),
non_cite_templates, non_cite_templates /
max(wikitext.revision.content_chars, 1), infobox_templates,
cn_templates + 1, cn_templates / max(wikitext.revision.content_chars, 1),
who_templates + 1, who_templates / max(wikitext.revision.content_chars, 1),
main_article_templates,
main_article_templates / max(wikitext.revision.content_chars, 1),
(english.stemmed.revision.stem_chars /
max(wikitext.revision.content_chars, 1)),
log(paragraphs_without_refs_total_length + 1), words_to_watch_count,
words_to_watch_count / max(wikitext.revision.words, 1), idioms_count,
idioms_count /
max(wikitext.revision.words, 1), words_to_watch_count + idioms_count,
(words_to_watch_count + idioms_count) / max(wikitext.revision.words, 1)
]
wp10 = wikipedia.article + local_wiki
"""
Based largely on work by Morten Warncke-Wang et al.[1] and with a few
improvements and extensions that Morten identified after publication.
1. Warncke-Wang, M., Cosley, D., & Riedl, J. (2013, August). Tell me more: An
actionable quality model for wikipedia. In Proceedings of the 9th
International Symposium on Open Collaboration (p. 8). ACM.
http://opensym.org/wsos2013/proceedings/p0202-warncke.pdf
proportion_of_badwords = portuguese.parent_revision.badwords / \
max(portuguese.parent_revision.words, 1)
proportion_of_misspellings = portuguese.parent_revision.misspellings / \
max(portuguese.parent_revision.words, 1)
proportion_of_informals = portuguese.parent_revision.informals / \
max(portuguese.parent_revision.words, 1)
added_badwords_ratio = proportion_of_badwords_added / \
max(proportion_of_badwords, 0.01)
added_misspellings_ratio = proportion_of_misspellings_added / \
max(proportion_of_misspellings, 0.01)
added_informals_ratio = proportion_of_informals_added / \
max(proportion_of_informals, 0.01)
damaging = util.no_lang_damaging + enwiki.badwords + enwiki.informals + [
log(portuguese.diff.badwords_added + 1),
log(portuguese.diff.badwords_removed + 1),
log(portuguese.diff.informals_added + 1),
log(portuguese.diff.informals_removed + 1),
log(portuguese.diff.misspellings_added + 1),
log(portuguese.diff.misspellings_removed + 1),
proportion_of_badwords_added,
proportion_of_badwords_removed,
proportion_of_informals_added,
proportion_of_informals_removed,
proportion_of_misspellings_added,
proportion_of_misspellings_removed,
added_badwords_ratio,
added_informals_ratio,
added_misspellings_ratio
]
proportion_of_informals_removed = spanish.diff.informals_removed / max(spanish.diff.words_removed, 1)
proportion_of_badwords = spanish.parent_revision.badwords / max(spanish.parent_revision.words, 1)
proportion_of_misspellings = spanish.parent_revision.misspellings / max(spanish.parent_revision.words, 1)
proportion_of_informals = spanish.parent_revision.informals / max(spanish.parent_revision.words, 1)
added_badwords_ratio = proportion_of_badwords_added / max(proportion_of_badwords, 0.01)
added_misspellings_ratio = proportion_of_misspellings_added / max(proportion_of_misspellings, 0.01)
added_informals_ratio = proportion_of_informals_added / max(proportion_of_informals, 0.01)
damaging = (
util.no_lang_damaging
+ enwiki.badwords
+ enwiki.informals
+ [
log(spanish.diff.badwords_added + 1),
log(spanish.diff.badwords_removed + 1),
log(spanish.diff.informals_added + 1),
log(spanish.diff.informals_removed + 1),
log(spanish.diff.misspellings_added + 1),
log(spanish.diff.misspellings_removed + 1),
proportion_of_badwords_added,
proportion_of_badwords_removed,
proportion_of_informals_added,
proportion_of_informals_removed,
proportion_of_misspellings_added,
proportion_of_misspellings_removed,
added_badwords_ratio,
added_informals_ratio,
added_misspellings_ratio,
]
proportion_of_badwords = vietnamese.parent_revision.badwords / \
max(vietnamese.parent_revision.words, 1)
proportion_of_misspellings = vietnamese.parent_revision.misspellings / \
max(vietnamese.parent_revision.words, 1)
proportion_of_informals = vietnamese.parent_revision.informals / \
max(vietnamese.parent_revision.words, 1)
added_badwords_ratio = proportion_of_badwords_added / \
max(proportion_of_badwords, 0.01)
added_misspellings_ratio = proportion_of_misspellings_added / \
max(proportion_of_misspellings, 0.01)
added_informals_ratio = proportion_of_informals_added / \
max(proportion_of_informals, 0.01)
damaging = util.no_lang_damaging + enwiki.badwords + enwiki.informals + [
log(vietnamese.diff.badwords_added + 1),
log(vietnamese.diff.badwords_removed + 1),
log(vietnamese.diff.informals_added + 1),
log(vietnamese.diff.informals_removed + 1),
log(vietnamese.diff.misspellings_added + 1),
log(vietnamese.diff.misspellings_removed + 1),
proportion_of_badwords_added,
proportion_of_badwords_removed,
proportion_of_informals_added,
proportion_of_informals_removed,
proportion_of_misspellings_added,
proportion_of_misspellings_removed,
added_badwords_ratio,
added_informals_ratio,
added_misspellings_ratio
]