You are viewing a single comment's thread from:

RE: The digital book of we

in #thoughts7 years ago

Yep, at this point in time, it is far from viable but potentially it could create a pretty decent view perhaps of our digital world at least. I am glad that a lot of my 'childhood days' were analogue ;D

Language processing has it so limitations.

LAnguage processing does but if possible to that point, how good would the ability to predict emotions considering the massive array of data from a wide range of other people as well as all the potential to analyse and match circumstance to reaction. a decade or two, who knows what an AI will be able to predict in our behaviour patterns. I'd say that in all likelihood, it would be more accurate than today's psychologists.

Would you be interested in hearing her opinion based on merely that?

I am likely to anyway by the looks.

Sort:  

Emotional prediction is possible already. Crackbook has ran tests on it as widely publicized, and so have insurance companies already with apps. Only very few were busted but it does happen and lots can be mined. They mostly happen based on shorter sentences/content tho and have been different in approach so far but the tech is evolving and only few years away. HR departments worldwide are looking forward to it.

With your content here I see problems because of the huge amount of paragraphs. There are too many if or elements in psychology but not sure that was the scope of the essay anyway.

Connecting the dots and everything... there’s already platforms that can do that. So aggregating everything in a story... with some targeted training Watson isn’t far of being capable of telling the story of the data’s made accessible to them.

With your content here I see problems because of the huge amount of paragraphs. There are too many if or elements in psychology but not sure that was the scope of the essay anyway.

I would be interested to see what would happen if there was 'Quotebook' that took excerpts the aI deemed important enough to quote :)

Connecting the dots and everything... there’s already platforms that can do that. So aggregating everything in a story... with some targeted training Watson isn’t far of being capable of telling the story of the data’s made accessible to them.

Yep, I am not saying it would give an accurate portrayal but it would definitely give an interesting view of a life.

I actually thought about the “quotebook” because that works rather well already but when it comes to providing psychological analysis that could lead to a totally different result.

Which is also why in criminal cases more often than not more than one specialist is asked to asses the subject.

It can massively distort the long term view but standalone it does work.

I should actually tell Jojo to make a Steem clone of it for your posts. The Reddit bot was rather enjoyed actually. :D