Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Content-Based Social
Recommendation with Poisson
Matrix Factorization
Eliezer de Souza da Silva
PhD student
Department of ...
2
Introduction
● Basic Problem: Recommendation of items to users given user interaction
with some items
User 1 User 2 User...
3
Challenges and opportunities in RS
Research
User 1 User 2 User 3
Item 1 Item 2 Item 3 Item 4 Item 5
Topics
4
Challenges and opportunities in RS
Research
User 1 User 2 User 3
Item 1 Item 2 Item 3 Item 4 Item 5
Topics
User 4
User
S...
5
Challenges and opportunities in RS
Research
● Incorporate
○ Social network analysis tools and methods
○ Content analysis...
6
Joint modelling of user social
network and item topic content
● User social network
○ Homophily
○ Item exposure positive...
7
[Topics, Words]
[Topics, Users]
[Items, Topics]
[Items, Words]
[Items, Users]
Observed
Latent
8
Poisson Matrix Factorization with Content and Social
trust information (PoissonMF-CS)
9
Items Topic Model
10
User preference and social factors
11
Poisson Matrix Factorization with Content and Social
trust information (PoissonMF-CS)
12
Inference
• Batch variational inference:
• Conjugate model with auxiliary variable “tricky” for each
Poisson likelihood...
13
Item Recommendations
• Top-M items for each user:
– Approximate expected value of user-item matrix for each
unseen item...
14
Application
Artist recommendation (Last-fm dataset):
• User-artist interactions counts
• User-user social network
• Art...
15
Results
● Avg. Recall Metric:
● Compare with previous work:
○ Collaborative Topic Regression (CTR)
○ Collaborative Topi...
16
Results
PoissonMF-CS (K =10) and
Gaussian-based models
PoissonMF-CS (K =10) and other Poisson
factorization models
17
Results
18
Conclusion
• Model including social and topic information in Poisson
matrix factorization using coupled latent factors
...
19
Questions?
https://github.com/zehsilva/poissonmf_cs
20
Content-based Social Poisson Factorization for recommendation
Upcoming SlideShare
Loading in …5
×
Upcoming SlideShare
What to Upload to SlideShare
Next
Download to read offline and view in fullscreen.

0

Share

Download to read offline

Content-Based Social Recommendation with Poisson Matrix Factorization (ECML-PKDD 2017)

Download to read offline

Presentation from ECML-PKDD 2017 of joint work with my supervisors Helge Langseth and Heri Ramampiaro, about a Poisson factorization model for recommendations that integrates topic models in the items set and social connections between users.

code: https://github.com/zehsilva/poissonmf_cs
preprint version: https://inajourney.files.wordpress.com/2012/11/poissonmf_cs.pdf

Related Books

Free with a 30 day trial from Scribd

See all

Related Audiobooks

Free with a 30 day trial from Scribd

See all
  • Be the first to like this

Content-Based Social Recommendation with Poisson Matrix Factorization (ECML-PKDD 2017)

  1. 1. Content-Based Social Recommendation with Poisson Matrix Factorization Eliezer de Souza da Silva PhD student Department of Computer Science, NTNU Joint work with Helge Langseth and Heri Ramampiaro ECML-PKDD 2017
  2. 2. 2 Introduction ● Basic Problem: Recommendation of items to users given user interaction with some items User 1 User 2 User 3 Item 1 Item 2 Item 3 Item 4
  3. 3. 3 Challenges and opportunities in RS Research User 1 User 2 User 3 Item 1 Item 2 Item 3 Item 4 Item 5 Topics
  4. 4. 4 Challenges and opportunities in RS Research User 1 User 2 User 3 Item 1 Item 2 Item 3 Item 4 Item 5 Topics User 4 User Social Network
  5. 5. 5 Challenges and opportunities in RS Research ● Incorporate ○ Social network analysis tools and methods ○ Content analysis (topic models, sentiment/intent/mood) ○ New rich contextual information ■ location, activity, user intent/goal, etc.
  6. 6. 6 Joint modelling of user social network and item topic content ● User social network ○ Homophily ○ Item exposure positively influenced by peers (positive “peer-pressure”) ● Item content analysis ○ Enrich items latent factors with topic model ○ Cold start items ○ Preferences can be influenced by topics
  7. 7. 7 [Topics, Words] [Topics, Users] [Items, Topics] [Items, Words] [Items, Users] Observed Latent
  8. 8. 8 Poisson Matrix Factorization with Content and Social trust information (PoissonMF-CS)
  9. 9. 9 Items Topic Model
  10. 10. 10 User preference and social factors
  11. 11. 11 Poisson Matrix Factorization with Content and Social trust information (PoissonMF-CS)
  12. 12. 12 Inference • Batch variational inference: • Conjugate model with auxiliary variable “tricky” for each Poisson likelihood term: • Running time for each iteration depends on the sparse observations: – O(K(obs_W + obs_R + obs_S + U + D + W ))
  13. 13. 13 Item Recommendations • Top-M items for each user: – Approximate expected value of user-item matrix for each unseen item for ranking Rud User preferences Shared item topic intensity Item topic offset Weighted sum of social network neighbors interactions with item
  14. 14. 14 Application Artist recommendation (Last-fm dataset): • User-artist interactions counts • User-user social network • Artist-tags counts Dataset size: – 1892 users, 17632 artists, 11946 tags – 25434 user–user connections, 92834 user–items interactions, and 186479 user–tag–items entries.
  15. 15. 15 Results ● Avg. Recall Metric: ● Compare with previous work: ○ Collaborative Topic Regression (CTR) ○ Collaborative Topic Regression with Social Matrix Factorization (CTR-SMF) ○ Collaborative topic Poisson factorization (CTPF) ○ Social Poisson Factorization (SPF)
  16. 16. 16 Results PoissonMF-CS (K =10) and Gaussian-based models PoissonMF-CS (K =10) and other Poisson factorization models
  17. 17. 17 Results
  18. 18. 18 Conclusion • Model including social and topic information in Poisson matrix factorization using coupled latent factors • Inference is computationally efficient with variational inference • Future work: – Non-negative relational learning – Non-parametric extensions – Scalable inference (SVI)
  19. 19. 19 Questions? https://github.com/zehsilva/poissonmf_cs
  20. 20. 20 Content-based Social Poisson Factorization for recommendation

Presentation from ECML-PKDD 2017 of joint work with my supervisors Helge Langseth and Heri Ramampiaro, about a Poisson factorization model for recommendations that integrates topic models in the items set and social connections between users. code: https://github.com/zehsilva/poissonmf_cs preprint version: https://inajourney.files.wordpress.com/2012/11/poissonmf_cs.pdf

Views

Total views

1,303

On Slideshare

0

From embeds

0

Number of embeds

889

Actions

Downloads

13

Shares

0

Comments

0

Likes

0

×