Text Moderation

If you allow users to post text, maybe in the form of comments on blogs, you are responsible for moderating and removing offensive, inappropriate and illegal content - all things which could hurt your brand if left unchecked.

​

You may not have the time to do it, so let our Text Moderator service be your frontline defence. Using the best of AI, we can process text in near real-time and identify potential problems, allowing you to decide what you want to do about it. It may just be a bit of light-hearted banter, or it may be something truly offensive - we can highlight it; you decide.

​

Our AI engine checks for the following:

• Identity attack

• Insult

• Obscenity

• Toxicity (rude, disrespectful, unreasonable)

• Severe toxicity (aggressive, hateful, very disrespectful)

• Threat

The categories are marked on a scale of 0 to 1, with 1 being the worst value. The system measures these against "reasonable" threshold values to arrive at a decision as to whether the text needs moderation. You are welcome to apply your own thresholds for your particular needs.

​

Additionally, the system checks for emojis, non-printing characters, and HTML, in case someone might be trying to "game" the system.