Try my new trust metric!

[ciphergoth]

I have implemented my new trust metric, and applied it (fairly crudely) to LiveJournal friends lists as an experiment. The result is a tool that tells you who is "closest" to your friends list who isn't actually on it, like "popwithfriends" but more sophisticated.

Who is closest to your friends list?

Let me know how plausible the results seem to you.

Update: Shit, I've broken the machine it was running on. Back later when I've made it check the load average before running the metric... Update: Back again now, with a rate limiter to keep the load average reasonable. Gosh, hadn't expected such popularity!

Comments

[flickgc.livejournal.com]

Coo! Purdy! Shiny!

Question... I get different results with your thing than with this.

I don't pretend for a minute to understand anything but the most superficial bits of the trust metrics (although I think it's a good idea), but I thought that - when it was run on the basis that it is here - it would behave in the same way as the LJ tool.

Sorry - that's a bit wibbly, but I think it makes sense.

So why's it different? Have I missed something? (probably!)

[ciphergoth.livejournal.com]

I'm using a different algorithm. The differences are, in short:

1) your votes for a person count more if you list fewer people

2) once you get listed, your friends list helps determine who might be listed in future. But not everyone's vote is equally valuable.

For a precise description, see trustmetrics.

Which do you think is more accurate for you?

[flickgc.livejournal.com]

Ah, right - the weighting makes sense, actually!

(I have read the full description... I just didn't get it. Sorry, dear.)

As to which is more accurate... It's tricky to say - your version is more noticably skewed by the cardboards than the LJ on (becuase they tend to have small friends lists, I guess).

Most of the people on them I know of. But I'm not entirely sure whether I would friend them, becuase I don't know them! Some of them are people I've deliberately not friended, but that's the same on both lists.

Sorry - not a lot of help, is it?

[rliz.livejournal.com]

> 1) your votes for a person count more if you list fewer people

Clevvver.

[gregoid.livejournal.com]

That is very clever giving more weight to the smaller lists. Did you also take into consideration the length of time the person has had their journal?