The comments in this dataset come from an archive of Wikipedia talk page comments. These have been annotated by Jigsaw for toxicity, as well as (for the main config) a variety of toxicity subtypes, including severe toxicity, obscenity, threatening language, insulting language, and identity attacks. This dataset is a replica of the data released for the Jigsaw Toxic Comment Classification Challenge and Jigsaw Multilingual Toxic Comment Classification competition on Kaggle, with the test dataset merged with the test_labels released after the end of the competitions. Test data not used for scoring has been dropped. This dataset is released under CC0, as is the underlying comment text.

To use this dataset:

import tensorflow_datasets as tfds

ds = tfds.load('wikipedia_toxicity_subtypes', split='train')
for ex in ds.take(4):
 print(ex)

See the guide for more informations on tensorflow_datasets.

Dataset Card for Jigsaw Toxic Comments

  Dataset





  Dataset Description

The Jigsaw Toxic Comments dataset is a benchmark dataset created for the Toxic Comment Classification Challenge on Kaggle. It is designed to help develop machine learning models that can identify and classify toxic online comments across multiple categories of toxicity.

Curated by: Jigsaw (a technology incubator within Alphabet Inc.) Shared by: Kaggle Language(s) (NLP): English License: CC0 1.0… See the full description on the dataset page: https://huggingface.co/datasets/anitamaxvim/jigsaw-toxic-comments.

Toxic-comments (Teeny-Tiny Castle)

This dataset is part of a tutorial tied to the Teeny-Tiny Castle, an open-source repository containing educational tools for AI Ethics and Safety research.

  How to Use

from datasets import load_dataset

dataset = load_dataset("AiresPucrs/toxic_content", split = 'train')

Description

Data from Toxic Comment Classification Challenge without modification

For using it in Jigsaw Rate Severity of Toxic Comments

Example usage: ☣️ Jigsaw - Super Simple Naive Bayes [LB=0.768]

Please, DO upvote if you use the dataset!

tillschwoerer/toxic-comments dataset hosted on Hugging Face and contributed by the HF Datasets community

Preporcessed Toxic Comments Classification Dataset

The obstacle I faced in Toxic Comments Classification Challenge was the preprocessing part. One can easily improve their LB performance if the preprocessing is done right.

This is the preprocessed version of Toxic Comments Classification Challenge dataset. The code for preprocessing: https://www.kaggle.com/fizzbuzz/toxic-data-preprocessing

At the end of 2017 the Civil Comments platform shut down and chose make their ~2m public comments from their platform available in a lasting open archive so that researchers could understand and improve civility in online conversations for years to come. Jigsaw sponsored this effort and extended annotation of this data by human raters for various toxic conversational attributes.

In the data supplied for this competition, the text of the individual comment is found in the comment_text column. Each comment in Train has a toxicity label (target), and models should predict the target toxicity for the Test data. This attribute (and all others) are fractional values which represent the fraction of human raters who believed the attribute applied to the given comment.

The data also has several additional toxicity subtype attributes. Models do not need to predict these attributes for the competition, they are included as an additional avenue for research. Subtype attributes are:

severe_toxicity obscene threat insult identity_attack sexual_explicit

Additionally, a subset of comments have been labelled with a variety of identity attributes, representing the identities that are mentioned in the comment. The columns corresponding to identity attributes are listed below. Only identities with more than 500 examples in the test set (combined public and private) will be included in the evaluation calculation. These identities are shown in bold.

male female transgender other_gender heterosexual homosexual_gay_or_lesbian bisexual other_sexual_orientation christian jewish muslim hindu buddhist atheist other_religion black white asian latino other_race_or_ethnicity physical_disability intellectual_or_learning_disability psychiatric_or_mental_illness other_disability

The Toxicity Dataset*by Surge AIThis dataset contains 500 toxic and 500 non-toxic comments from a variety of popular social media platforms. Rather than operating under a strict definition of toxicity, we asked our team to identify comments that they personally found toxic. ## Columns text: the text of the comment* is_toxic: whether or not the comment is toxic

tcapelle/jigsaw-toxic-comment-classification-challenge dataset hosted on Hugging Face and contributed by the HF Datasets community

Dataset

This dataset was created by chinmay Das

Contents

About this dataset

Trained models to predict toxic comments on all 3 Jigsaw Toxic Comment Challenges. https://github.com/unitaryai/detoxify

How to use

import transformers

# original
model_path = "../input/detoxify/original"
self.tokenizer = transformers.BertTokenizer.from_pretrained(model_path)
self.encoder = transformers.BertForTokenClassification.from_pretrained(
  f"{model_path}/pytorch_model.bin",
  config = transformers.BertConfig.from_pretrained(f"{model_path}/config.json")
).bert

# unbiased
model_path = "../input/detoxify/unbiased"
self.tokenizer = transformers.RobertaTokenizer.from_pretrained(model_path)
self.encoder = transformers.RobertaForSequenceClassification.from_pretrained(
  f"{model_path}/pytorch_model.bin",
  config = transformers.RobertaConfig.from_pretrained(f"{model_path}/config.json")
).roberta

# multilingual
model_path = "../input/detoxify/multilingual"
self.tokenizer = transformers.XLMRobertaTokenizer.from_pretrained(model_path)
self.encoder = transformers.XLMRobertaForSequenceClassification.from_pretrained(
  f"{model_path}/pytorch_model.bin",
  config = transformers.XLMRobertaConfig.from_pretrained(f"{model_path}/config.json")
).roberta

About Dataset This is a hand-labelled toxicity data set containing 1000 comments crawled from YouTube videos about the Ferguson unrest in 2014. In addition to toxicity, this data set contains labels for multiple subclassifications of toxicity which form a hierarchical structure. Each comment can have multiple of these labels assigned. The structure can be seen in the following enumeration:

*IsToxic - IsAbusive IsThreat IsProvocative IsObscene - IsHatespeech IsRacist IsNationalist IsSexist IsHomophobic IsReligiousHate - IsRadicalism

Original Data Source: Youtube toxic comments

Processed Jigsaw Toxic Comments Dataset

This is a preprocessed and tokenized version of the original Jigsaw Toxic Comment Classification Challenge dataset, prepared for multi-label toxicity classification using transformer-based models like BERT. ⚠️ Important Note: I am not the original creator of the dataset. This dataset is a cleaned and restructured version made for quick use in PyTorch deep learning models.

  📦 Dataset Features

Each example contains:

text: The… See the full description on the dataset page: https://huggingface.co/datasets/Koushim/processed-jigsaw-toxic-comments.

Dataset

This dataset was created by AllHailSammy

Released under GPL 2

Contents

UIT-ViCTSD (Vietnamese Constructive and Toxic Speech Detection) is a dataset for constructive and toxic speech detection in Vietnamese. It consists of 10,000 human-annotated comments.

akcit-ijf/jigsaw-toxic-comment-train-processed-seqlen128 dataset hosted on Hugging Face and contributed by the HF Datasets community

Dataset

This dataset was created by trandong2932002

Released under MIT

Contents

This text dataset includes 3,481 social media user comments posted in response to political news posts and videos on Twitter, YouTube, and Reddit in August, 2021. The dataset also includes MTurk workers’ annotations of these comments as hateful, offensive, and/or toxic; and codes assigned by researchers describing various rhetorical dimensions of these comments.

This classifier is a fine-tuned Roberta-base model using the simpletransformers toolkit for classifying toxic language. We use the Kaggle Toxic Comments Challenge dataset as training data. This dataset contains Wikipedia comments classified as toxic or non-toxic.

Dataset

This dataset was created by Rowan Curry

Contents