Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API

kolinko.github.io

Post reply on HN
Print view View original

61–70 of 94 posts

Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API

#62

Post by taylorbuley »

I'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 analysis of myself than through empiricism. My online self may not be my "true self" but it certainly represents a portion of who I want to be.

Now to contextualize, I wonder how we trend together and apart from the median user model as individuals and the community? The distributions seem interesting -- for example at a glance we appear heavy on challenge seekers but light on stability!

Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API

#63

Post by minimaxir »

sysk wrote: post #55
edw519 wrote: post #53

This is hilarious! I'm ranked first in: Cheerfulness (Before or after morning coffee?) Orderliness (Because of my 3rd normal form sock drawer?) Gregariousness (Before of after my 3rd beer?) Agreeableness (I disagree! Watson needs debugging.) I'm ranked dead last in: Imagination (No one I know could imagine how this could be.) Authority-challenging (My teachers & bosses would disagree.) Intellect (Before or after my m…

This could make a good movie plot. The all powerful AI analyses your characteristics, gets it all wrong and assigns you to the wrong line of work (Harry Potter?) :)

This happens in the pilot episode of Futurama.

Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API

#65

Post by bane »

kolinko wrote: post #22
bane wrote: post #18

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 interact with one another, so when I ask somebody, "do you think is an agreeable person?" and they reply "sure I think he is!" they're consulting that model to provide me an answer. Humans can even do a kind of pairwise sorting on that model a…

Well, 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 the results mean exactly and how solid they are.

Well, in a sense, most of us all only know each other through our comments, and that's all we can ever base an assessment like this on. By proxy we have to assume that when people's inner thoughts leak out into the Internet on a forum like this (and in a sustained enough way to make them a top-100 karma earner) that their aggregate corpus of comments will be a reasonable insight into who they are.

So for all purposes that you, I or Watson can demonstrate, "do you think X is Y" and "do you think X's comments show Y" are functionally the same.

edit

I just checked what Watson thinks are my needs. Apparently I don't have many, and everybody on HN has an extreme need for Challenge.

I almost feel like these results require a lot of interpretation, and that interpretation is about as reliable as a horoscope.

Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API

#66

Post by fivedogit »

rebelde wrote: post #60
fivedogit wrote: post #51

Earlier quoted context omitted.

Hi. I built a Chrome extension for Hacker News that lets people follow others and get notifications when they are replied to or their karma changes. http://hackbook.club The ext basically matches my original vision now, so I'm thinking hard about what features to add next, but I'm not quite sure I understand what you're asking for. Are you saying that when the extension says "So and so replied to you", you want, say,…

On 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 (2) shows the category if that HNer is in the top or bottom 10% in any category. For instance: edw519 33 minutes ago | link Top: Cheerfulness, Orderliness, Gregariousness, Agreeableness Bottom: Imagination, Authority-challenging, Intellect (I'…

Ok this actually makes a lot of sense. When someone replies to you on Facebook, you generally know who they are. On Hacker News, you have no idea. Including some basic (albeit imperfect) information along with a notification reply could be really useful.

This could even be true for the follow-feed mechanism. If you're following someone on Twitter or friends with them on Facebook and they write something that appears in your feed, you are already familiar with the author on some level. In the Hackbook extension, it's fairly common to follow people and not know much about them at all. Again, including some basic Watson-generated information along with the "a user you're following wrote a comment" newsfeed item could be helpful.

Let me give it some more thought and maybe I'll have time to work on it this week.

Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API

#69

Post by bashinator »

minimaxir wrote: post #63
sysk wrote: post #55

Earlier quoted context omitted.

This could make a good movie plot. The all powerful AI analyses your characteristics, gets it all wrong and assigns you to the wrong line of work (Harry Potter?) :)

This happens in the pilot episode of Futurama.

I thought the joke was that the AI was completely correct.

Re: Show HN: Comments by Top HN Posters Analysed by IBM's Watson User Modelling API

#70

Post by d0vs »

rebelde wrote: post #26

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 traits are outliers from the norm. You get insulted by somebody ranked low in Sympathy? No need to worry. I'm serious. Somebody do this. It would look great on the résumé.

> You get insulted by somebody ranked low in Sympathy? No need to worry.

Because you totally need to worry if someone insults you on the Internet.

Post reply on HN