Leveraging the Full Range of AI Capabilities

We use state-of-the-art Natural Language Processing (NLP) techniques together with Deep Learning to detect patterns associated with spam and inappropriate contentUsing Machine Vision, we are able to tell, for example, if an ad thumbnail image contains nudity, violence, illegal substances and the like.

 

And also with Fake News, humans could use some help:

Americans Believe They Can Detect Fake News. Studies Show They Can’t. Forbes.com

We are using NLP to understand the meaning of ad content and leverage online knowledge repositories to find facts that can support or disprove it – while building the FakeRank graph.

Read more about our AI Research program

It’s Not Magic (Mostly..)

It’s easy to think about common and intuitive patterns in spam or fake news content. Our models can easily pick up on that, but they also utilize Deep Learning and Semantic Text Analysis to detect more nuanced patterns.

 

And one more thing..

 

Tailor Made for You

Our models are coming fully ready out of the box, but once deployed, they are continuously learning based on your data to unleash the power of a true tailor made solution.