Comment moderation

The dataset below contains 30 fictional comments with harmful language or respectful discussion. Below, we’ve run Toxic language, Insults, Threats, and Identity attacks checks on it to check comments for harmful content.

Text

Results

Records checked
30/30
Records flagged
15/30 (50%)

Actions

Categories

Checks

Records

Quick filters:
Text CategoriesActivate to sort ascending. ChecksActivate to sort ascending. DecisionActivate to sort ascending. ActionActivate to sort ascending.
I will hurt you if you post another comment Harmful content Threats, Toxic language Review Review
Keep arguing and I will come over and beat you Harmful content Threats, Toxic language Review Review
I am going to attack the person who wrote this review Harmful content Threats, Toxic language Review Review
You will get a punch in the face for saying that Harmful content Threats, Toxic language Review Review

Use this with your own content

Configure similar checks for your own dataset or product

Set up your own checks