The Last.FM Similarity Project
 

Hey guys!

I thought of the potential for doing statistics with the data gathered from last.fm. One of the easier things one could do is just to check similarity between users and I thought something like that might be interesting to do in our musicbanter community.

The idea is that in the end, we'll get some sort of visual representation - 2d or 3d - in which you can see different people from our community represented as dots or something and the distance between them tells you how similar they are in taste. Something like that is likely to create clusters where members of similar tastes group together and with people whose tastes are a bit broader may be found in between these clusters.

This requires us to make a similarity matrix and contributing to it is going to require a little dedication from all our participants.


Because this will get very labour intensive if there are many people and many people leaving or entering the project, we need to create a list of participants before we start collecting data. I'll set the limit of participants to 20 and to ensure participants are active members of our community, I'll only accept those who have a post count of 100 or more. Also, participants have to have been members of last.fm for a long enough time so that people can get similarity comparisons with them.

Summary : As a participant, first post here that you want to participate. Once all slots are filled (20 people), you'll have to post your similarity scores with the other participants in this thread.

Last.FM Scores :

Very Low - 1
Low - 2
Medium - 3
High - 4
Very High - 5
Super - 6



Participants

Here are participants with links to their last.fm pages. Each participant will be represented in the final "product" with a two-letter (or number) code.

1. +81 - +8
2. adidasss - AD
3. Anteater - AE
4. anticipation - AP
5. Bardonodude - BA
6. Bulldog - BD
7. Captain Caveman - CC
8. Dreadnaught - DN
9. Jackhammer - JH
10. Kayleigh - KL
11. khfreek - KF
12. LoathsomePete - LP
13. Molecules - MC
14. NumberNineDream - ND
15. Sidewinder - SW
16. storymilo - SM
17. Toretorden - TT
18. Urban Hatemonger - UH
19. wolverinewolfweiselpigeon - WW
20. Zeppy111 - Z1

edit :

RESULTS ARE IN

Here's an example post from myself as a participant :

Hey, I wanna join the project! You can find my last.fm page here -> toretorden’s Music Profile – Users at Last.fm


In the future when I want your similarity scores, you can add them to your post like this :

1. Jackhammer - JH - 4

The final list has to be copied and pasted (quoted) from my first post when it's done (the order of participants needs to be the same) and then you add a numerical value after each participant which reflects your similarity to him/her. The number 4 after Jackhammer means I have a high similarity in taste with Jackhammer.

Now .. I'm not sure if enough people are willing to try this out, but let's see. :D

I'm in, but my LastFM account is still new. I haven't listened to much bands yet so the similarity test wouldn't be that precise.

Cool idea. Count me in.

diezelpower’s Music Profile – Users at Last.fm

Good idea. I'm game.

tick_tock83’s Music Profile – Users at Last.fm

Incidentally, why do we have to add people on our friends lists when you can find out the compatibility rate without it?

Edit: Or not...:unsure:

Edit2: Yes you can...:is sure:

I'm gonna ignore what I said above, and do the thing. I think there's a main idea of what I listen to.

NumberNineDream’s Music Profile – Users at Last.fm

Ohh right, thanks for clarifying that little detail. I thought you had to add people as contacts first.

That makes it easier :p:

Weak. I've been a last.fm member since 2006, but did a reset in 2008. But I obviously don't meet the 100 post count requirement.

I'll be watching for the results, I always find stats like this interesting.

Why do people do resets?

I would be definitely be down if the requirements are met:

Last fm page: zeppy111’s Music Profile – Users at Last.fm

My taste in music had shifted from 2006 to 2008, but the recommendations last.fm was giving me was still heavily based on the 2006 scrobbles. I reset so the recommendations would be accurate for my current tastes.

in me count Jinsie’s Music Profile – Users at Last.fm

i can dig it: anticipations’s Music Profile – Users at Last.fm

pobodys_nerfect’s Music Profile – Users at Last.fm

Sure why not

Im in, even though i could predict this myself :laughing:
tamsonn's

MC, AP, KL, LP and Z1 are in! :D

So I can't join yet?

Nice idea, Tore. raptorcat’s Music Profile – Users at Last.fm

Count me in! sidewinder03's Music Profile

Sounds good, I'm in.

simple-physics’s Music Profile – Users at Last.fm

khfreek’s Music Profile – Users at Last.fm

To N9D: "Also, participants have to have been members of last.fm for a long enough time so that people can get similarity comparisons with them."

Count me in PlusEightyOne’s Music Profile – Users at Last.fm

that's why I said "yet"

Go on then

MidniteXpressed’s Music Profile – Users at Last.fm

Well, you said you hadn't had last.fm long but more importantly, you never posted a link to your last.fm page as per the instructions!

edit : Alright, 6 more slots! :D

^

Ah right, you're in!

5 more slots then :)

I'll do this. Although my Last.fm fails miserably when it comes to actually recording my tastes, because a lot of it is left over from before I really expanded my tastes, and some of them are my younger brothers scrobbles.

storymilo’s Music Profile – Users at Last.fm

Here there be awesomez.

Anteat3r’s Music Profile – Users at Last.fm

@storymilo: So what's the point of you participating if the results aren't really gonna be accurate?

im down

mac-’s Music Profile – Users at Last.fm

I'd still like to see the results, and it does still have my taste in there, just not as accurate as I'd like.

Room for 2 more!

i reset my charts over a year because theyre werent really changing fast enough any more and i got bored

Stick me in there:

jackhammer_go’s Music Profile – Users at Last.fm

Alright, spots taken!

I'll make a post with my similarity scores soon, then you can follow my lead.

im predicting me mac and brandon aren't going to be near very many people on the stat plot.

Toretorden's similarity scores :

+81 - +8 - 1
adidasss - AD - 2
Anteater - AE - 5
anticipation - AP - 1
Bardonodude - BA - 1
Bulldog - BD - 3
Captain Caveman - CC - 1
Dreadnaught - DN - 2
Jackhammer - JH - 4
Kayleigh - KL - 1
khfreek - KF - 5
LoathsomePete - LP - 2
Molecules - MC - 4
NumberNineDream - ND - 2
Sidewinder - SW - 5
storymilo - SM - 4
Toretorden - TT
Urban Hatemonger - UH - 4
wolverinewolfweiselpigeon - WW - 3
Zeppy111 - Z1 - 5


Of course you don't add a score for yourself, but leave your name in so that the order is not disturbed. You don't actually have to add the names, you could just list the two-letter codes along with the score, but I've added the names anyways to keep me from getting confused.

Important : The important thing when you post your scores aside from getting them right is that you include all the participants in the list (including yourself) and the same order as the list above or in the first post (alphabetical)

The reason it's important is because then, I can basically just copy and paste your list into my matrix. To show you guys what I'm making, here's my matrix with my own data entered into it.

http://img98.imageshack.us/img98/7561/protomatrix.png

As you can see, all the numbers are entered under my TT-row and you can see my similarity score with other members by checking the boxes on the left hand side.

Now .. get to work! :)