MMCF: Multimodal Collaborative Filtering
for Automatic Playlist Continuation
Team ‘hello world!’ (2nd place), main track
Hojin Yang*, Yoonki Jeong, Minjin Choi, and Jongwuk Lee
Sungkyunkwan University, Republic of Korea
2
Motivation & Challenges
Automatic Playlist Continuation
➢Dataset: Million Playlist Dataset (MPD)
3
Playlist title
Tracks
in the
playlist
Metadata of tracks
(artist, album)
Challenge Set
4
1 2 3 4 5 6 7 8 9 10
# of tracks 0 1 5 10 5 10 25 100 25 100
Title
available
Yes Yes Yes Yes No No Yes Yes Yes Yes
Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled
# of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000
Few tracks in the
first part
Many tracks in the
first part
Many tracks in the
random position
Challenge Set
5
1 2 3 4 5 6 7 8 9 10
# of tracks 0 1 5 10 5 10 25 100 25 100
Title
available
Yes Yes Yes Yes No No Yes Yes Yes Yes
Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled
# of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000
No tracks in the playlist
How to deal with an edge case?
Challenge Set
6
1 2 3 4 5 6 7 8 9 10
# of tracks 0 1 5 10 5 10 25 100 25 100
Title
available
Yes Yes Yes Yes No No Yes Yes Yes Yes
Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled
# of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000
Scarce information
How to treat playlists with scarce
information?
Challenge Set
7
1 2 3 4 5 6 7 8 9 10
# of tracks 0 1 5 10 5 10 25 100 25 100
Title
available
Yes Yes Yes Yes No No Yes Yes Yes Yes
Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled
# of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000
Various types of Input
How to deal with various types of
input?
8
Proposed Model
Overview of the Proposed Model
➢An ensemble method with two components.
◆ Autoencoder for tracks and metadata for tracks.
◆ CharCNN for playlist titles.
9
Overview of the Proposed Model
➢An ensemble method with two components
◆ Autoencoder for tracks and metadata for tracks
◆ CharCNN for playlist titles
10
Autoencoder
11
encoder decoder
Compressed representation
of the input image
28X28X3 pixels
➢Learn a latent representation by reconstructing the
input itself
28X28X3 pixels
Collaborative Autoencoder
12
1
0
1
1
0
1Hey Jude
Rehab
Yesterday
Dancing Queen
Mamma Mia
Viva la Vida
encoder decoder
➢Learn a latent representation of a given playlist
consisting of a set of tracks
0.9
0.01
0.78
0.9
0. 6
0.8
✔
Top-1
Recommendation
Output
Collaborative Autoencoder
13
1
0
1
1
0
1
1
0
1
0
0
0
0.9
0.01
0.78
0.9
0. 6
0.8Hey Jude
Rehab
Yesterday
Dancing Queen
Mamma Mia
Viva la Vida
dropout
encoder decoder
1
0
1
1
0
1
➢Training with dropout
◆ Some positive input values are corrupted (set to zero).
Collaborative Autoencoder
14
➢Training with dropout
◆ Some positive input values are corrupted (set to zero).
How to utilize the metadata such
as artists and albums?
1
0
1
1
0
1
1
0
1
0
0
0
0.9
0.01
0.78
0.9
0. 6
0.8Hey Jude
Rehab
Yesterday
Dancing Queen
Mamma Mia
Viva la Vida
dropout
encoder decoder
1
0
1
1
0
1
Utilizing Metadata
15
1
0
1
1
0
1
0
1
1
0
0.9
0.01
0.78
0.9
0. 6
0.8
0.2
0.98
0.9
0.6
Hey Jude
Rehab
Yesterday
Dancing Queen
Mamma Mia
Viva la Vida
encoder decoder
➢Simply concatenate an artist vector corresponding
to the track vector.
➢Randomly choose either the playlist or its artists as
input.
16
Training Strategy: Hide-and-Seek
Training Strategy: Hide-and-Seek
➢Randomly choose either the playlist or its artists as
input.
17
Training Strategy: Hide-and-Seek
➢Randomly choose either the playlist or its artists as
input.
18
➢About 0.5~9.0% improvement over the baseline
19
sequential shuffled
Input 1 5 10 25 100 25 100
Track 0.111 0.153 0.183 0.215 0.159 0.306 0.301
Track + Artist 0.121 0.156 0.184 0.216 0.172 0.317 0.303
Gain(%) +9.00 +1.96 +0.55 +0.47 +8.18 +3.60 +0.66
* R-precision
Training Strategy: Hide-and-Seek
➢How to design a dropout strategy?
20
1
0
1
1
0
1
0
1
1
0
1
0
1
0
0
0
0
0
0
0
0.9
0.23
0.78
0.1
0.1
0.8
0.2
0.98
0.9
0.1
Hey Jude
Rehab
Yesterday
Dancing Queen
Mamma Mia
Viva la Vida
dropout
encoder decoder
1
0
1
1
0
1
0
1
1
0
Training Strategy: Dropout Scheme
Hide
Training Strategy: Dropout Scheme
➢Dropout ratios vary in the input size.
21
1 2 3 4 5 6 7 8 9 10
# of tracks 0 1 5 10 5 10 25 100 25 100
Title
available
Yes Yes Yes Yes No No Yes Yes Yes Yes
Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled
# of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000
Dropout a lot
Dropout a little
➢Dropout can be sequential or random.
22
1 2 3 4 5 6 7 8 9 10
# of tracks 0 1 5 10 5 10 25 100 25 100
Title
available
Yes Yes Yes Yes No No Yes Yes Yes Yes
Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled
# of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000
Dropout
the back part
Dropout
randomly
Training Strategy: Dropout Scheme
CharCNN for Playlist Titles
➢An ensemble method with two components
◆ Autoencoder for tracks and metadata for tracks
◆ CharCNN for playlist titles
23
Character-level CNN for NLP
➢Effective for capturing spatial locality of a
sequence of texts
24
I like this
song
very
much
0.1 0.3 0.2 0.6
0.2 0.6 -1.2 -0.2
-2.1 0.2 0.1 0.4
-2.1 0.9 -3.1 1.4
0.1 0.3 -0.2 0.1
0.4 0.1 0.7 0.1
I
like
this
song
very
Filter (3 by k )
2.2
2.3
-1.3
0.9
max
pooling
Conv layer
2.3
Feature
much
k-dimension embedding
convolutional
Character-level CNN for NLP
➢Effective for capturing spatial locality of a
sequence of texts
25
I like this
song
very
much
0.1 0.3 0.2 0.6
0.2 0.6 -1.2 -0.2
-2.1 0.2 0.1 0.4
-2.1 0.9 -3.1 1.4
0.1 0.3 -0.2 0.1
0.4 0.1 0.7 0.1
I
like
this
song
very
Filters (3 by k )
convolutional
2.2
2.3
-1.3
0.9
max
pooling
2.3
Feature
much
k-dimension embedding
Conv layer
2.2
2.3
-1.3
0.9
Conv layers
2.2
2.3
-1.3
0.9
1.2
2.4
-1.1
0.4
max
pooling
2.3
1.2
2.4
Feature vector
convolutional
CharCNN for Playlist Titles
➢The playlist title is represented by a short text, implying
the abstract description of a playlist.
➢Leverage character-level embedding instead of
word-level embedding.
26
Conv layers
Feature vector
Combining Two Models
➢Simplest method: 𝑤𝒊𝒕𝒆𝒎 = 0.5 and 𝑤𝒕𝒊𝒕𝒍𝒆 = 0.5
27
Combining Two Models
➢The accuracy of the AE relies on the number of
tracks in a playlist.
◆ Dynamic: Set weights according to the number of items.
28
Items
Playlist Title
Chill songs
0.7
0.4
0.9
0.1
0.2
0.1
0.2
0.3
0.7
0.1
0.6
0.4
0.7
0.2
0.2
AE
CNN
𝑤_𝑖𝑡𝑒𝑚 = 5
𝑤_𝑡𝑖𝑡𝑙𝑒 = 1
Combining Two Models
➢The weight of two models is combined
dynamically depending on the size of input.
➢The playlist title is primarily useful for playlists with
very few tracks.
29
sequential shuffled
Title leverage 0 1 5 10 25 100 25 100
Item only - 0.121 0.156 0.184 0.216 0.172 0.317 0.303
Title only 0.078 - - - - - - -
Constant(0.5) 0.078 0.131 0.158 0.181 0.211 0.161 0.310 0.292
Dynamic 0.078 0.131 0.158 0.184 0.216 0.172 0.317 0.303
Gain(%) - - - +1.6 +2.3 +6.8 +2.2 +3.7
* R-precision
Q&A
30
Thank you!
Hojin Yang
- undergraduate student, SKKU
- code: github.com/hojinYang/spotify_recSys_challenge_2018
- email: hojin.yang7@gmail.com

MMCF: Multimodal Collaborative Filtering for Automatic Playlist Conitnuation

  • 1.
    MMCF: Multimodal CollaborativeFiltering for Automatic Playlist Continuation Team ‘hello world!’ (2nd place), main track Hojin Yang*, Yoonki Jeong, Minjin Choi, and Jongwuk Lee Sungkyunkwan University, Republic of Korea
  • 2.
  • 3.
    Automatic Playlist Continuation ➢Dataset:Million Playlist Dataset (MPD) 3 Playlist title Tracks in the playlist Metadata of tracks (artist, album)
  • 4.
    Challenge Set 4 1 23 4 5 6 7 8 9 10 # of tracks 0 1 5 10 5 10 25 100 25 100 Title available Yes Yes Yes Yes No No Yes Yes Yes Yes Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled # of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 Few tracks in the first part Many tracks in the first part Many tracks in the random position
  • 5.
    Challenge Set 5 1 23 4 5 6 7 8 9 10 # of tracks 0 1 5 10 5 10 25 100 25 100 Title available Yes Yes Yes Yes No No Yes Yes Yes Yes Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled # of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 No tracks in the playlist How to deal with an edge case?
  • 6.
    Challenge Set 6 1 23 4 5 6 7 8 9 10 # of tracks 0 1 5 10 5 10 25 100 25 100 Title available Yes Yes Yes Yes No No Yes Yes Yes Yes Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled # of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 Scarce information How to treat playlists with scarce information?
  • 7.
    Challenge Set 7 1 23 4 5 6 7 8 9 10 # of tracks 0 1 5 10 5 10 25 100 25 100 Title available Yes Yes Yes Yes No No Yes Yes Yes Yes Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled # of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 Various types of Input How to deal with various types of input?
  • 8.
  • 9.
    Overview of theProposed Model ➢An ensemble method with two components. ◆ Autoencoder for tracks and metadata for tracks. ◆ CharCNN for playlist titles. 9
  • 10.
    Overview of theProposed Model ➢An ensemble method with two components ◆ Autoencoder for tracks and metadata for tracks ◆ CharCNN for playlist titles 10
  • 11.
    Autoencoder 11 encoder decoder Compressed representation ofthe input image 28X28X3 pixels ➢Learn a latent representation by reconstructing the input itself 28X28X3 pixels
  • 12.
    Collaborative Autoencoder 12 1 0 1 1 0 1Hey Jude Rehab Yesterday DancingQueen Mamma Mia Viva la Vida encoder decoder ➢Learn a latent representation of a given playlist consisting of a set of tracks 0.9 0.01 0.78 0.9 0. 6 0.8 ✔ Top-1 Recommendation Output
  • 13.
    Collaborative Autoencoder 13 1 0 1 1 0 1 1 0 1 0 0 0 0.9 0.01 0.78 0.9 0. 6 0.8HeyJude Rehab Yesterday Dancing Queen Mamma Mia Viva la Vida dropout encoder decoder 1 0 1 1 0 1 ➢Training with dropout ◆ Some positive input values are corrupted (set to zero).
  • 14.
    Collaborative Autoencoder 14 ➢Training withdropout ◆ Some positive input values are corrupted (set to zero). How to utilize the metadata such as artists and albums? 1 0 1 1 0 1 1 0 1 0 0 0 0.9 0.01 0.78 0.9 0. 6 0.8Hey Jude Rehab Yesterday Dancing Queen Mamma Mia Viva la Vida dropout encoder decoder 1 0 1 1 0 1
  • 15.
    Utilizing Metadata 15 1 0 1 1 0 1 0 1 1 0 0.9 0.01 0.78 0.9 0. 6 0.8 0.2 0.98 0.9 0.6 HeyJude Rehab Yesterday Dancing Queen Mamma Mia Viva la Vida encoder decoder ➢Simply concatenate an artist vector corresponding to the track vector.
  • 16.
    ➢Randomly choose eitherthe playlist or its artists as input. 16 Training Strategy: Hide-and-Seek
  • 17.
    Training Strategy: Hide-and-Seek ➢Randomlychoose either the playlist or its artists as input. 17
  • 18.
    Training Strategy: Hide-and-Seek ➢Randomlychoose either the playlist or its artists as input. 18
  • 19.
    ➢About 0.5~9.0% improvementover the baseline 19 sequential shuffled Input 1 5 10 25 100 25 100 Track 0.111 0.153 0.183 0.215 0.159 0.306 0.301 Track + Artist 0.121 0.156 0.184 0.216 0.172 0.317 0.303 Gain(%) +9.00 +1.96 +0.55 +0.47 +8.18 +3.60 +0.66 * R-precision Training Strategy: Hide-and-Seek
  • 20.
    ➢How to designa dropout strategy? 20 1 0 1 1 0 1 0 1 1 0 1 0 1 0 0 0 0 0 0 0 0.9 0.23 0.78 0.1 0.1 0.8 0.2 0.98 0.9 0.1 Hey Jude Rehab Yesterday Dancing Queen Mamma Mia Viva la Vida dropout encoder decoder 1 0 1 1 0 1 0 1 1 0 Training Strategy: Dropout Scheme Hide
  • 21.
    Training Strategy: DropoutScheme ➢Dropout ratios vary in the input size. 21 1 2 3 4 5 6 7 8 9 10 # of tracks 0 1 5 10 5 10 25 100 25 100 Title available Yes Yes Yes Yes No No Yes Yes Yes Yes Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled # of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 Dropout a lot Dropout a little
  • 22.
    ➢Dropout can besequential or random. 22 1 2 3 4 5 6 7 8 9 10 # of tracks 0 1 5 10 5 10 25 100 25 100 Title available Yes Yes Yes Yes No No Yes Yes Yes Yes Track order Seq Seq Seq Seq Seq Seq Seq Seq Shuffled Shuffled # of playlists 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 1,000 Dropout the back part Dropout randomly Training Strategy: Dropout Scheme
  • 23.
    CharCNN for PlaylistTitles ➢An ensemble method with two components ◆ Autoencoder for tracks and metadata for tracks ◆ CharCNN for playlist titles 23
  • 24.
    Character-level CNN forNLP ➢Effective for capturing spatial locality of a sequence of texts 24 I like this song very much 0.1 0.3 0.2 0.6 0.2 0.6 -1.2 -0.2 -2.1 0.2 0.1 0.4 -2.1 0.9 -3.1 1.4 0.1 0.3 -0.2 0.1 0.4 0.1 0.7 0.1 I like this song very Filter (3 by k ) 2.2 2.3 -1.3 0.9 max pooling Conv layer 2.3 Feature much k-dimension embedding convolutional
  • 25.
    Character-level CNN forNLP ➢Effective for capturing spatial locality of a sequence of texts 25 I like this song very much 0.1 0.3 0.2 0.6 0.2 0.6 -1.2 -0.2 -2.1 0.2 0.1 0.4 -2.1 0.9 -3.1 1.4 0.1 0.3 -0.2 0.1 0.4 0.1 0.7 0.1 I like this song very Filters (3 by k ) convolutional 2.2 2.3 -1.3 0.9 max pooling 2.3 Feature much k-dimension embedding Conv layer 2.2 2.3 -1.3 0.9 Conv layers 2.2 2.3 -1.3 0.9 1.2 2.4 -1.1 0.4 max pooling 2.3 1.2 2.4 Feature vector convolutional
  • 26.
    CharCNN for PlaylistTitles ➢The playlist title is represented by a short text, implying the abstract description of a playlist. ➢Leverage character-level embedding instead of word-level embedding. 26 Conv layers Feature vector
  • 27.
    Combining Two Models ➢Simplestmethod: 𝑤𝒊𝒕𝒆𝒎 = 0.5 and 𝑤𝒕𝒊𝒕𝒍𝒆 = 0.5 27
  • 28.
    Combining Two Models ➢Theaccuracy of the AE relies on the number of tracks in a playlist. ◆ Dynamic: Set weights according to the number of items. 28 Items Playlist Title Chill songs 0.7 0.4 0.9 0.1 0.2 0.1 0.2 0.3 0.7 0.1 0.6 0.4 0.7 0.2 0.2 AE CNN 𝑤_𝑖𝑡𝑒𝑚 = 5 𝑤_𝑡𝑖𝑡𝑙𝑒 = 1
  • 29.
    Combining Two Models ➢Theweight of two models is combined dynamically depending on the size of input. ➢The playlist title is primarily useful for playlists with very few tracks. 29 sequential shuffled Title leverage 0 1 5 10 25 100 25 100 Item only - 0.121 0.156 0.184 0.216 0.172 0.317 0.303 Title only 0.078 - - - - - - - Constant(0.5) 0.078 0.131 0.158 0.181 0.211 0.161 0.310 0.292 Dynamic 0.078 0.131 0.158 0.184 0.216 0.172 0.317 0.303 Gain(%) - - - +1.6 +2.3 +6.8 +2.2 +3.7 * R-precision
  • 30.
    Q&A 30 Thank you! Hojin Yang -undergraduate student, SKKU - code: github.com/hojinYang/spotify_recSys_challenge_2018 - email: hojin.yang7@gmail.com