We created i am safe online to give everyone an opportunity to create a safe, abuse-free online environment anywhere they want.
The service analyses input text and decides whether its content is abusive or not. It uses a sophisticated machine learning algorithm to evaluate how abusive a message can be. We trained this algorithm on a human-curated dataset of real comments from various sources.
The analysis can then be used by you to keep your website, blog or comment section “hate” free.
Currently, the service can only indicate if the content is abusive or not, but shortly we’re planning to improve it by using more topic-specific data for training. This will allow i am safe online to understand the context of the message better and identify phrases that can be offensive to a specific community.
We tried to make it as easy to integrate as possible, and we would love you to try it out and give us some feedback!