1. 1 Learning Relevance from Heterogeneous Social Network and Its Application in Online Targeting Chi Wang (UIUC), Rajat Raina (FB), David Fong (Stanford), Ding Zhou (FB), Jiawei Han (UIUC), Greg Badros (FB)

2. 2 Facebook Social Graph

3. 3 Relevance Matters

4. 4 Opportunity Can we use the rich, social information flowing through online social network to personalize user recommendations?

5. 5 Ad Targeting Goal: Estimate CTR for a given user/ad pair. Importance For users: relevant information For advertisers: more consumers

6. 6 Process Overview

7. 7 Concept Extraction (CE)

8. 8 Concept Aggregation and Match

9. 9 Baseline for CA and CM

10. 10 Limitations Concept Aggregation –Different source type should have different weights. E.g., greetings between friends may not be as useful as clicks on a previous ad. Concept Match –Different concept type should have different weights. E.g., top level node Society VS. leaf node Society/Relationships/Dating. –One user concept can match multiple ad concepts. E.g., the video game FIFA Series can match the computer game or the sport soccer –User concept space and ad concept space are not necessarily identical.

11. 11 Solution: Idea1

12. 12 Solution: Idea2

13. 13 Proposed Better Model

14. 14 Algorithms-Gradient Method

15. 15 Experiments

16. 16 Logloss-gain

17. 17 Interpretations

18. 18 Conclusion