Модель Похожих товаров и её приложения

Айгуль Газимова (Joom.ru, Junior Machine Learning Developer) @ Moscow Python №78

"Похожие товары – это важная часть для любой e-commerce платформы, а особенно для маркетплейсов. Они помогают покупателям найти лучший товар за меньшие деньги. Joom – международная группа e-commerce и финтех компаний, которая работает как с B2B, так и с B2C сегментами с одной основной командой разработки поиска. Мы поговорим про:
- построение Модели Похожих Товаров для Joom Marketplace (B2C-продукт);
- разработку Модели Похожих Товаров для JoomPro (B2B-продукт), где у нас мало пользовательской истории;
- как с помощью такой модели мы улучшили алгоритмы рекомендаций".

Видео: https://moscowpython.ru/meetup/78/similar-products-search/

Transcript

1. SIMILAR PRODUCTS MODEL AND ITS’ APPLICATIONS Aigul Gazimova
2. Aliexpress SHEIN Similar Products Model Amazon
3. Joom Search Team ML engineers, Software engineers, Analytics
4. About me AIGUL GAZIMOVA Bachelor’s degree in applied mathematics and computer science Cloud storage → eCommerce Search@Joom
5. JoomPro platform for crossborder wholesale trade International group of e-commerce and fintech companies Joom Marketplace platform for shopping from all over the world Onfy pharmaceutical marketplace in Germany Joompay fintech service for daily financial transactions in Europe Joom Logistics business that provides logistics, technology and infrastructure services for crossborder eCommerce
6. Similar Products Model
7. Similar products: idea of using Word2Vec device_id product_id1 … product_idn
8. . . . device_id1 product_id1 … product_idn1 device_id2 product_id1 … product_idn2 device_idm product_id1 … product_idnm Similar products: idea of using Word2Vec
9. Word2Vec: recap
10. Center Center embedding and context embedding
11. V(KINGS) - V(KING) ≈ V(QUEENS) - V(QUEEN) Interpretation of embeddings
12. Issue: new products
13. Issue: new products
14. Image embeddings
15. Customer experience
16. Customer experience
17. Customer experience
18. Customer experience
19. How to build the Similar Products Model if we lack of user activity history?
20. How to build the Similar Products Model if we lack of user activity history? - word2vec approach
21. How to build the Similar Products Model if we lack of user activity history? - word2vec approach
22. How to build the Similar Products Model if we lack of user activity history? - word2vec approach - image-based embeddings
23. How to build the Similar Products Model if we lack of user activity history? - word2vec approach - image-based embeddings
24. How to build the Similar Products Model if we lack of user activity history? - word2vec approach - image-based embeddings to use content information for word2vec approach and to train the model on joom users’ history
25. category id product name from EUR 12 Content Similar Products Model: product features
26. Optimus Prime!!!
27. category_id product name BOW BPE category emb name emb Transformer encoder center embedding context embedding product_id category_id BOW BOW emb categories emb + Content Similar Products Model: architecture
28. category_id product name BOW BPE category emb name emb Transformer encoder center embedding context embedding product_id category_id BOW BOW emb categories emb + Content Similar Products Model: applications
29. Category embeddings: 2D representation with TSNE
30. Clusterization
31. • Similar Products Model for B2C • Content Similar Products Model for B2B • Why it’s useful in e-commerce platforms Conclusion
32. AIGUL GAZIMOVA jesaig@joom.com
33. BTW, we are hiring!