Putting in the system reliability for you prediction, some sort of a percentage out of 100 would at least give a disclaimer for your viewers that the predictions being given by your system isn't a 100% guaranteed price movement.
By using word "prediction", IMO almost all people realize it's not accurate 100%.
Or values are adjusted by placing external influences such as news, events, updates, hardforks, and other effects then determining the proportion of each effect in the prediction.
I'd be surprised if they did that. It means they already have the capabilities to allow a machine to judge the value of each news. Maybe it's not that far, I suspect they'll keywords and then categorize each keyword into a value. For example, "broke the channel up", "optimist", and so on into 1, and bear-related keywords like "dumped" and so on to negative on, and so on.
That would be sick.
That would involve Natural Language Processing (specifially Sentiment analysis) which AFAIK is harder topic than machine learning.
But the source news would greatly affect the prediction, Bitcoin/Cryptocurrency media usually use positive words while other media usually use negative words.