um-hn
AI
personality analysis
SHOW HN2014
180points
90comments
#122of 2,272 in 2014
Top 6%of 2014 by points
How HN took it
MixedNo clear lean
10 positive, 22 neutral or asking, 0 mixed, 17 critical, out of 49 comments about the project. Thread score -0.12.
Performance: 3 for, 7 againstIdea: 5 for, 3 againstExecution: 0 for, 4 againstDocs: 2 for, 0 againstDesign: 0 for, 2 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Fun tool to explore commenter personalities
From 2 comments
Very 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…
noxtras on HNI'm so incredibly grateful to have this user model https://pbs.twimg.com/media/B4xNHXtCAAAoFsw.png:large 2144 days into this experiement that is HN, I couldn't ask for a better…
taylorbuley on HN
Idea for browser extension using the data
From 2 comments
Somebody please make a browser plugin that uses this data. I know the data isn't perfect, but it would be nice to be able see who in a thread is a top HNer and which character…
rebelde on HNOn one level it could be a simple bookmarket (bookmark with JS). When clicked, it goes through the HN discussion page. When it finds a top HNer's username, it (1) colors it and…
rebelde on HN
Objections
Watson scores are not validated and may be meaningless
From 3 comments
One of the difficulties this kind of output has is that it runs very quickly into issues of semantics and model labeling. We humans each build models kind of like these when we…
bane on HNThese results look pretty spotty...the rankings seem about as random as picking out names from a Bingo machine. User ssciafani is among the top users in Cautiousness, Openness,…
danso on HNYeah, we couldn't find any proof that the results are meaningful in any way, but I think it's a fun tool nevertheless. One thing I know for sure is that Watson doesn't recognise…
kolinko on HN
Small sample comments give unreliable rankings
From 2 comments
Does not seem correct. The top poster for practicality has some useless thank you posts that provide nothing useful to the reader. StackOverflow even tends to lock questions that…
lnanek2 on HNThat's really cool. For some reason, the ranking changes every time I click on the same category. Edit: I just realised it's probably because the position of users who have the…
sysk on HN
Rankings change on repeated clicks due to ties
From 2 comments
You may want to consider using a stable sort to list the results. Users with equal scores are shuffled each time I click the topic name.
mgraczyk on HNThat's really cool. For some reason, the ranking changes every time I click on the same category. Edit: I just realised it's probably because the position of users who have the…
sysk on HN
Requests
Documentation of methodology and what scores mean
From 3 comments
I had a small giggle at my rankings, not going to lie. Without a definition for what each of these things are, it's a little unclear what this is actually saying. I'm assuming…
steveklabnik on HNWell, the real question this tool answers is not really "do you think X is Y", but "do you think X's comments show Y". There's also a lack of documentation from IBM as to what…
kolinko on HN> I almost feel like these results require a lot of interpretation, and that interpretation is about as reliable as a horoscope. I got a similar feeling from this - that's why…
kolinko on HN
Browser extension showing traits next to usernames
From 2 comments
Somebody please make a browser plugin that uses this data. I know the data isn't perfect, but it would be nice to be able see who in a thread is a top HNer and which character…
rebelde on HNOn one level it could be a simple bookmarket (bookmark with JS). When clicked, it goes through the HN discussion page. When it finds a top HNer's username, it (1) colors it and…
rebelde on HN
Fix broken navigation and history support
From 2 comments
Purely in terms of the website, you have no back button from the list of users but also no support for browser history. Otherwise cool!
xixixao on HNYou need to load your js files off https, by defalt your page does not work and you need to disable mixed content warnings for anything to load
dylz on HN
Questions
How were the scores calculated and what do they mean
From 2 comments
I had a small giggle at my rankings, not going to lie. Without a definition for what each of these things are, it's a little unclear what this is actually saying. I'm assuming…
steveklabnik on HNIBM's/Watson's documentation is really vague on what these things are, but there's some explaination to that in Wikipedia:…
kolinko on HN
Why are low-comment users ranked highly
From 1 comment
That's really cool. For some reason, the ranking changes every time I click on the same category. Edit: I just realised it's probably because the position of users who have the…
sysk on HN
Alternatives named
Twitter personality research via Economist article
From 1 comment
I first read about the "Big 5 personality Traits" in this Economist article last year. Very interesting read on how some researchers were using Twitter writings (specifically…
nnain on HN