Unraveling Multimodality with Large Language Models.pdf
ย
Improving Music Recommendation in Session-Based Collaborative Filtering by using Temporal Context
1. Improving Music Recommendation in
Session-Based Collaborative Filtering by using
Temporal Context
Ricardo Dias, Manuel J. Fonseca
University of Lisbon
Instituto Superior Tรฉcnico
ICTAI 2013
8. 1. Few work to study the usage of temporal
context combined with CF algorithms
2. Some take advantage of the songs listened in
sessions (and implicit context they provide)
3. None of them used temporal features
extracted from the sessions
9. 1. Few work to study the usage of temporal
context combined with CF algorithms
2. Some take advantage of the songs listened in
sessions (and implicit context they provide)
3. None of them used temporal features
extracted from the sessions
10. 1. Few work to study the usage of temporal
context combined with CF algorithms
2. Some take advantage of the songs listened in
sessions (and implicit context they provide)
3. None of them used temporal features
extracted from the sessions
14. User 1
Session 1
Session 2
โฆ
User 2 โฆ
Artist 1, Song 1, Timestamp
Artist 2, Song 3, Timestamp
Artist 1, Song 2, Timestamp
Artist 3, Song 5, Timestamp
Artist 4, Song 4, Timestamp
Time
16. Takes into account the songs a user has listened
and rated
Song 1
Song 2
Song 3
Song 4
Song 5
Alice
5
3
4
4
?
User 1
3
1
2
3
3
User 2
4
3
4
3
5
User 3
3
3
1
5
4
User 4
1
5
5
2
1
17.
18. Session profiles instead of User Profiles
Song 1
Song 2
Session 1
1
4
Session 2
1
Session 3
[Park2011]
Song 4
Song 5
4
?
2
1
Session 4
Session 5
Song 3
2
1
1
2
1
1
1
19. Song 1
Song 2
Session 1
1
4
Session 2
1
Session 3
Song 4
Song 5
4
?
2
1
Session 4
Session 5
Song 3
2
1
1
2
1
1
1
20. ๐ ๐ ,๐ = ๐๐ +
๐ฃ โ ๐(๐ ) ๐
๐๐๐ ๐, ๐ โ (๐ ๐ฃ,๐ โ ๐ ๐ฃ )
๐ฃ โ ๐(๐ ) ๐
๐๐๐(๐, ๐)
๐บ(๐) ๐ represents the k session neighbors for the active sessions, ๐ ๐ and ๐ ๐ the average
rating for the sessions s and v under comparison
23. ๐ ๐ ,๐ = ๐๐ +
๐ฃ โ ๐(๐ ) ๐
๐๐๐ ๐, ๐ โ (๐ ๐ฃ,๐ โ ๐ ๐ฃ )
๐ฃ โ ๐(๐ ) ๐
๐๐๐(๐, ๐)
๐บ(๐) ๐ represents the k session neighbors for the active sessions, ๐ ๐ and ๐ ๐ the average
rating for the sessions s and v under comparison
24. ๐ ๐ ,๐ = ๐๐ +
๐ฃ โ ๐(๐ ) ๐
๐๐๐ ๐, ๐ โ (๐ ๐ฃ,๐ โ ๐ ๐ฃ )
๐ฃ โ ๐(๐ ) ๐
๐๐๐(๐, ๐)
๐บ(๐) ๐ represents the k session neighbors for the active sessions, ๐ ๐ and ๐ ๐ the average
rating for the sessions s and v under comparison
31. Documents are collections of words
Each document is represented as a mixture of
latent topics, and each topic has probabilities of
generating various words
32. Uses Session Data Model with LDA
โ Documents: sessions
โ Words: songs
Sessions (Documents) are treated as Bag-ofSongs (Words)
33. Topic
Distribution
S1 and S2 more similar than S1 and S3 or S2 and
S3
1
0,8
0,6
0,4
Session 1
0,2
Session 2
Session 3
0
1
2
3
Topics
4
sim(s1, s2) = 0.163
sim(s1, s3) = 1.139
sim(s2, s3) = 0.871
44. HR@n ๏ Hit Ratio at N
(top-n recommendations)
MRR ๏ Mean Reciprocal Rank
45. Dataset โ Session split
Training: all the songs in sessions, except the last
one
Test: the last song in the session
46. Queries ๏ 1000 Sessions randomly
chosen
For i = 1
to i = 20
For each query obtained the top-100,
and recorded the presence and rank
given by the algorithms
56. Temporal Context improves recommendation
accuracy in Session-based CF
โ LDA based algorithm with best results
HR increased >200%
57. Combine temporal properties with other
context features
โ Example: locations, activities, user behavior
(skips), etc.
Conduct a user study by applying the algorithm
in a real world application