Anonymize Large-scale Sparse User Features at LINE Corp

2019/3/7 Machine Learning Production Pitch #1
Yeo Chaerim

Transcript

1. ANONYMIZE LARGE-SCALE SPARSE USER FEATURES AT LINE CORP CHAERIM YEO, LINE CORPORATION MACHINE LEARNING PRODUCTION PITCH #1, 2019/03/07
2. ABOUT ME l Chaerim Yeo(呂 彩林) l 2018.12 ~ LINE Corporation l Account Platform Development Dept. l Ad performance optimization
3. Agenda • Z-Features • Y-Features • Evaluation • Conclusion
4. Z-FEATURES
5. WHAT ARE Z-FEATURES
6. WHAT ARE Z-FEATURES
7. WHAT ARE Z-FEATURES
8. WHAT ARE Z-FEATURES
9. WHAT ARE Z-FEATURES
10. BENEFIT OF Z-FEATURES Reusable Flexible
11. LIMITATION OF Z-FEATURES Human Interpretable Extremely Sparse
12. Y-FEATURES
13. BEYOND Z-FEATURES Obfuscation Dimensionality Reduction
14. BEYOND Z-FEATURES Obfuscation Dimensionality Reduction With keeping information as far as possible
15. BEYOND Z-FEATURES Obfuscation Dimensionality Reduction SCDV https://arxiv.org/abs/1612.06778
16. OVERVIEW OF SCDV
17. INTEGRATE Z-FEATURES WITH SCDV
18. SYSTEM OVERVIEW
19. EVALUATION
20. DATA DIMENSION RELATIVE TO Z-FEATURES (LOG-SCALE) 0.0001 0.0010 0.0100 0.1000 1.0000 10.0000 100.0000 type1 type2 type3 type4 type5 type6 type7 type8 type9
21. DATA DENSITY LOG-SCALE 0.0000001 0.0000010 0.0000100 0.0001000 0.0010000 0.0100000 0.1000000 1.0000000 type1 type2 type3 type4 type5 type6 type7 type8 type9 z-features y-features
22. DATA SIZE RELATIVE TO Z-FEATURES 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 type1 type2 type3 type4 type5 type6 type7 type8 type9
23. USER DEMOGRAPHICS ESTIMATION MATRICS (RELATIVE TO Z-FEATURES) 0.95 0.96 0.97 0.98 0.99 1.00 1.01 1.02 gender age-group region precision recall f1-score
24. USER DEMOGRAPHICS ESTIMATION RUNNING TIME (RELATIVE TO Z-FEATURES) 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 gender age-group region training prediction
25. CONCLUSION
26. CONCLUSION l Anonymize user features based on SCDV l Enough to use in ML l Future works l Add workflow to production l Apply further dimensionality reduction l Auto encoders, PCA, …
27. THANK YOU