I don't see how they can be sure based on their criteria that those are fraudulent clicks. They could be web crawlers, users with various ad blocking tech, and

I don't see how they can be sure based on their criteria that those are fraudulent clicks. They could be web crawlers, users with various ad blocking tech, and so on. Going straight to fraud is premature.