o Ni$@sdZddlmZddlmZddlmZddlZddlZddlmm Z ddl m Z dZdZdZd Zd Zgd ZGd d d e jjZGddde jjZdS)z=CivilComments from Jigsaw Unintended Bias Kaggle Competition.)absolute_import)division)print_functionNa @article{DBLP:journals/corr/abs-1903-04561, author = {Daniel Borkan and Lucas Dixon and Jeffrey Sorensen and Nithum Thain and Lucy Vasserman}, title = {Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classification}, journal = {CoRR}, volume = {abs/1903.04561}, year = {2019}, url = {http://arxiv.org/abs/1903.04561}, archivePrefix = {arXiv}, eprint = {1903.04561}, timestamp = {Sun, 31 Mar 2019 19:01:24 +0200}, biburl = {https://dblp.org/rec/bib/journals/corr/abs-1903-04561}, bibsource = {dblp computer science bibliography, https://dblp.org} } a9 This version of the CivilComments Dataset provides access to the primary seven labels that were annotated by crowd workers, the toxicity and other tags are a value between 0 and 1 indicating the fraction of annotators that assigned these attributes to the comment text. The other tags are only available for a fraction of the input examples. They are currently ignored for the main dataset; the CivilCommentsIdentities set includes those labels, but only consists of the subset of the data with them. The other attributes that were part of the original CivilComments release are included only in the raw data. See the Kaggle documentation for more details about the available features. The comments in this dataset come from an archive of the Civil Comments platform, a commenting plugin for independent news sites. These public comments were created from 2015 - 2017 and appeared on approximately 50 English-language news sites across the world. When Civil Comments shut down in 2017, they chose to make the public comments available in a lasting open archive to enable future research. The original data, published on figshare, includes the public comment text, some associated metadata such as article IDs, timestamps and commenter-generated "civility" labels, but does not include user ids. Jigsaw extended this dataset by adding additional labels for toxicity and identity mentions. This data set is an exact replica of the data released for the Jigsaw Unintended Bias in Toxicity Classification Kaggle challenge. This dataset is released under CC0, as is the underlying comment text. z The CivilComments set here includes all the data, but only the basic seven labels (toxicity, severe_toxicity, obscene, threat, insult, identity_attack, and sexual_explicit). z The CivilCommentsIdentities set here includes an extended set of identity labels in addition to the basic seven labels. However, it only includes the subset (roughly a quarter) of the data with all these features. zchttps://storage.googleapis.com/jigsaw-unintended-bias-in-toxicity-classification/civil_comments.zip)malefemale transgender other_gender heterosexualhomosexual_gay_or_lesbianbisexualother_sexual_orientation christianjewishmuslimhindubuddhistatheistother_religionblackwhiteasianlatinoother_race_or_ethnicityphysical_disability#intellectual_or_learning_disabilitypsychiatric_or_mental_illnessother_disabilitycs eZdZdZfddZZS)CivilCommentsConfigz"Configuration for `CivilComments`.cs(tt|j||tjdd||_dS)Nz1.0.0)name descriptionversion)superr__init__tfdscoreVersioninclude_identity_labels)selfrrr& __class__[/home/ubuntu/.local/lib/python3.10/site-packages/tensorflow_datasets/text/civil_comments.pyr"js  zCivilCommentsConfig.__init__)__name__ __module__ __qualname____doc__r" __classcell__r*r*r(r+rgsrc@sLeZdZdZededdededdgZddZdd Z d d Z d d Z dS) CivilCommentsaClassification and tagging of 2M comments on news sites. This version of the CivilComments Dataset provides access to the primary seven labels that were annotated by crowd workers, the toxicity and other tags are a value between 0 and 1 indicating the fraction of annotators that assigned these attributes to the comment text. The other tags are only available for a fraction of the input examples. They are currently ignored for the main dataset; the CivilCommentsIdentities set includes those labels, but only consists of the subset of the data with them. The other attributes that were part of the original CivilComments release are included only in the raw data. See the Kaggle documentation for more details about the available features. F)rrr&CivilCommentsIdentitiesTcCsZdtji}gd}|jjr|t7}|D]}tj||<qtjj |t tj |ddt dS)Ntext)toxicitysevere_toxicityobscenethreatinsultidentity_attacksexual_explicit)r3r4zOhttps://www.kaggle.com/c/jigsaw-unintended-bias-in-toxicity-classification/data)builderrfeaturessupervised_keyshomepagecitation) r#r<Textbuilder_configr&IDENTITY_LABELStffloat32r$ DatasetInfo_COMMON_DESCRIPTION FeaturesDict _CITATION)r'r<labelslabelr*r*r+_infos  zCivilComments._infocCs|t}tjjtjjtj |dd|j j ddtjjtjj tj |dd|j j ddtjjtjj tj |dd|j j ddgS)zReturns SplitGenerators.z train.csvtarget)filenametoxicity_labelr&)r gen_kwargsztest_public_expanded.csvr4ztest_private_expanded.csv)download_and_extract _DOWNLOAD_URLr#r$SplitGeneratorSplitTRAINospathjoinrAr& VALIDATIONTEST)r' dl_managerdl_pathr*r*r+_split_generatorss.      zCivilComments._split_generatorscCsTi}|d|d<t|||d<|D]}||s|tvrdSt||||<q|S)N comment_textr3r4)floatrB)r'rowrN other_labelsexamplerJr*r*r+_parse_row_as_examples z#CivilComments._parse_row_as_examplec cs~gd}|r |t7}tjj|#}t|}|D]}||||}|r,|d|fVqWddS1s8wYdS)aYields examples. Each example contains a text input followed by several toxicity subtype scores and identity labels if include_identity_labels is True. Args: filename: the path of the file to be read for this split. toxicity_label: indicates 'target' or 'toxicity' to capture the variation in the released labels for this dataset. include_identity_labels: Whether to include the additional extended set of identity labels. In this case, only those examples with values for all labels are yielded. Yields: A dictionary of features, all floating point except the input text. )r5r6r7r8r9r:idN)rBrCiogfileGFilecsv DictReaderrb) r'rMrNr&rIfreaderr_rar*r*r+_generate_exampless "z CivilComments._generate_examplesN) r,r-r.r/r_CC_DESCRIPTION_CC_IDENTITIES_DESCRIPTIONBUILDER_CONFIGSrKr\rbrkr*r*r*r+r1ps" ' r1)r/ __future__rrrrgrUtensorflow.compat.v2compatv2rCtensorflow_datasets.public_api public_apir#rHrFrlrmrQrBr$ BuilderConfigrGeneratorBasedBuilderr1r*r*r*r+s