A Multilabel Classification Problem wherein a comment is classified into 3 labels : Toxicity, Insult & Threat.

The dataset can be found at : https://www.kaggle.com/c/jigsaw-unintended-bias-in-toxicity-classification/data

We have one single csv file for training.

Columns in train data: Comment_text: This is the data in string format which we have to use to find the toxicity. target: Target values which are to be predicted (has values between 0 and 1)

Data also has additional toxicity subtype features:

• severe_toxicity
• obscene
• threat
• insult
• identity_attack
• sexual_explicit

Comment_text data also has identity attributes carved out from it, some of which are:

• male
• female
• homosexual_gay_or_lesbian
• christian
• jewish
• muslim
• black
• white
• asian
• latino
• psychiatric_or_mental_illness

Apart from above features the train data also provides meta-data from jigsaw like:

• toxicity_annotator_count
• identity_anotator_count
• article_id
• funny
• sad
• wow
• likes
• disagree
• publication_id
• parent_id
• article_id
• created_date