• Share
  • Email
  • Embed
  • Like
  • Save
  • Private Content
Contemplated these 3 Flows for Playlist Similarity
 

Contemplated these 3 Flows for Playlist Similarity

on

  • 468 views

 

Statistics

Views

Total Views
468
Views on SlideShare
460
Embed Views
8

Actions

Likes
0
Downloads
0
Comments
0

5 Embeds 8

http://roelandp.posterous.com 3
http://www.lmodules.com 2
http://www.slideshare.net 1
http://www.linkedin.com 1
https://www.linkedin.com 1

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Contemplated these 3 Flows for Playlist Similarity Contemplated these 3 Flows for Playlist Similarity Document Transcript

    • 3 flow considerations about implementing similar playlists
    • WOULD NOT WORK: compile the top X (eg. 10) list of artists of each playlist sort this top X list by artist_ID compare two of these sorted lists with Pearson Correlation (=r) r can range from -1 to 1, where 0 is no linear correlation and 1 & -1 are perfectly linear if(abs(r)) = close to 1, high linear correlation Wouldnt work because, if artist ID's differ just one, still there would be high linear correlation, but music completely different
    • WOULD WORK: compile the top X (eg. 10) list of artists of each playlist compare the X members of List A to the X members of List B foreach exact hit, add a +1 to the score score can range from 0 to X, where 0 is no similarity and X is perfectly similar if(score = close to X) we have a very similar list a lists based upon absolute matching artists.
    • WOULD WORK ADVANCED: compile the top X (eg. 10) list of artists of each playlist retrieve single Tag for each artist (genre) compare the X Tags of List A to the X Tags of List B foreach exact hit, add a +1 to the score score can range from 0 to X, where 0 is no similarity and X is perfectly similar if(score = close to X) we have a very similar list in this case, lists based upon similar music styles, instead of absolute matching artists.