猜你喜歡 – 打造高度擴展的個人化電商推薦

Transcript

2. 猜你喜歡 – 打造高度擴展的 個人化電商推薦
3. 01 02 03 04 Multi-Stage Recommender Retrieval Ranking Challenges in LINE SHOPPING CONTENT 05 Re-rank 06 Model Training
4. Arthur Huang LINE Taiwan Machine Learning Engineer Work Experience • LINE Taiwan MLE (2021~Now) • SHOPLINE DE (2019~2021)
5. Challenges in LINE SHOPPING SECTION 01
6. Challenges in LINE SHOPPING 特點項目文字 特點項目 999 特點項目文字 特點項目 Complex Scenario • More than 20 types of recommendations.
7. Challenges in LINE SHOPPING 特點項目文字 特點項目 999 特點項目文字 特點項目 User Product >millions > ten millions Computational Explosion
8. SECTION 02 Multi-Stage Recommender
9. Multi-Stage Recommender Item Corpus • Quickly retrieve users' interested items. Ranking Re-rank millions hundreds dozens dozens Recommended Items • Ranking based on user behavior in the module. Ranking by Diversity, Freshness Business Logic. Retrieval • Ranking by Diversity, Freshness, Business Logic millions hundreds dozens dozens
10. SECTION 03 Retrieval Quickly retrieve users' interested items.
11. Retrieval - Training Two-Tower Model • Learning User-Item Embeddings • Target • Positive：Clicked Items • Negative：In-batch negative sampling
12. In-Batch Negative Sampling Click Click Click . . . User Ite m Positive Item
13. In-Batch Negative Sampling Click Click Click . . . User Ite m Negative Item
14. Feature Engineering Example : Spotify Million Playlist Dataset
15. Feature Engineering Example : Spotify Million Playlist Dataset • Numeric Feature • Normalization • Power Transform • CTR • Categorical Feature • One-Hot Encoding • Label Encoding + Embedding Layer • e.g. User ID, Item ID • Feature Hashing • Ordinal Encoding • Frequency Encoding • Text Feature • Bert Encoding
16. Quickly retrieval user's interested items. Retrieval - Inference
17. Online Serving Retrieval - Inference Online Offline
18. Item2Item Retrieval - Inference
19. SECTION 04 Ranking Ranking based on user behavior in the module.
20. Ranking - Training Deep Ranking Network • Learning the probability of click event. • Target (Focus on Module Interaction) • Positive : Click • Negative : Impression but no click. Ranking based on user behavior in the module.
21. Ranking - Target Positive Negative Negative Negative
22. Why can't we use items that were impression but not clicked as negative samples during retrieval? Item Corpus Ranking Re-rank millions hundreds dozens dozens Recommended Items Retrieval • Interest: Click • No Interest: Almost Item Corpus • Very Interest: Click • Interest: Impression but not Click
23. Ranking - Inference Distributed Model Inference
24. SECTION 05 Re-rank Ranking by Diversity, Freshness, Business Logic
25. Rerank Diversity Freshness • Do not show items form the same category in a sequence. • Promote fresher items. Business Logic • Promotion / Holiday Campagin • Product Profit
26. Does re-rank cause the metrics to decrease?