Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API
1–10 of 94 posts
Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API
#2Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API
#3Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API
#4Any feedback? :)
Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API
#5Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API
#6Very nice project, thanks for introducing me to IBM Watson. I will also play with it a little bit:) I don't know how it determines Openness and stuff, pasted one of my texts in there, it gave out 'interesting' results:)
One thing I know for sure is that Watson doesn't recognise irony - I posted this poem and it was marked as "cheerful" and "optimistic": http://bukowski.net/poems/a_smile_to_remember.php
Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API
#7Any feedback? :)
Selfishly, it would also be pretty cool to be able to select a person and see their rankings across the categories :-)
Oh, and your "top 100 users" link on the front page is broken as it's not an absolute link.
Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API
#8Any feedback? :)
Otherwise cool!
Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API
#9Any feedback? :)
Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API
#10Any feedback? :)
I'd be interested to learn what some of the headings really mean. For example, the only listing I'm near the top of is "Trust" (I'm #4) but I don't really know what that means in this context. I had a quick look at the IBM docs but couldn't figure it out. Selfishly, it would also be pretty cool to be able to select a person and see their rankings across the categories :-) Oh, and your "top 100 users" link on the fron…
- rankings across all the categories. good idea - shouldn't be hard to do, but don't have the time to implement it today :)
- as for what the headings mean, from what I understand they are related to http://en.wikipedia.org/wiki/Big_Five_personality_traits - but yes, the IBM watson docs are very vague on this subject