true Bayesian estimate

This forum is for help with this web site. Please read the FAQ and then search this forum before asking questions here; most likely the question you are about to ask has already been answered.

true Bayesian estimate

Postby Dan42 » Wed Jul 02, 2003 12:42 pm

I think the top 10% vids should be sorted according to a true Bayesian estimate, like IMDb.com, i.e.

weighted rank (WR) = (v ÷ (v+m)) × R + (m ÷ (v+m)) × C

where:
R = average for the movie (mean) = (Rating)
v = number of votes for the movie = (votes)
m = minimum votes required to be listed in the top 10% (3)
C = the mean vote across the whole report (all vids)

It would give a more stable list, with newer vids having to work their way up (as they should) instead of appearing high in the list and then going down as more people add opinions.
Dan42
Encyclopedist & PHP/SQL guy @ Anime News Network
Webmaster @ Dan42's Jin-Roh Lair & Chiyoko, Millennium Actress
User avatar
Dan42
 
Joined: 22 Jun 2001

Re: true Bayesian estimate

Postby Farlo » Wed Jul 02, 2003 2:29 pm

Dan42 wrote:I think the top 10% vids should be sorted according to a true Bayesian estimate, like IMDb.com, i.e.

weighted rank (WR) = (v ÷ (v+m)) × R + (m ÷ (v+m)) × C

where:
R = average for the movie (mean) = (Rating)
v = number of votes for the movie = (votes)
m = minimum votes required to be listed in the top 10% (3)
C = the mean vote across the whole report (all vids)

It would give a more stable list, with newer vids having to work their way up (as they should) instead of appearing high in the list and then going down as more people add opinions.


gah..math/any derivative bad...*STAB*

ok enough of my idiocy there, interesting idea, but i like it the way it is...more of a game of chance
User avatar
Farlo
 
Joined: 02 Apr 2002
Location: Fort Smith, Arkansas
Status: old timer amver

Postby Phade » Wed Jul 02, 2003 4:49 pm

Hey Dan42,

That is an excellent suggestion! I will fiddle with it and see what results come up. ^_^

Phade.
User avatar
Phade
Site Admin
 
Joined: 20 Oct 2000
Location: Little cabin in the woods...

Postby EarthCurrent » Wed Jul 02, 2003 5:21 pm

Yeah, that does provide a far more stable curve.
Good idea.
User avatar
EarthCurrent
 
Joined: 12 Feb 2002

Postby dwchang » Wed Jul 02, 2003 5:56 pm

EarthCurrent wrote:Yeah, that does provide a far more stable curve.
Good idea.


I bet my vids go bye-bye :(. Not that it matters though.

Anyway, I also agree, I think a more stable scoring system is needed. Currently there are *too many* videos on there with only 3 ratings and too many videos (IMO) that don't belong there. Again...IMO...please no flames.
-Daniel
Newest Video: Through the Years and Far Away aka Sad Girl in Space
User avatar
dwchang
Sad Boy on Site
 
Joined: 04 Mar 2002
Location: Madison, WI

Postby dwchang » Wed Jul 02, 2003 6:07 pm

Phade: Can you possibly post the results of your findings with this equation? I'm curious to see the distribution as well.
-Daniel
Newest Video: Through the Years and Far Away aka Sad Girl in Space
User avatar
dwchang
Sad Boy on Site
 
Joined: 04 Mar 2002
Location: Madison, WI


Return to Site Help & Feedback

Who is online

Users browsing this forum: No registered users and 0 guests