Zilliz logo
user icon

Vipshop's Personalized Recommender System

Vipshop is a leading online discount retailer for brands in China. The company offers high quality and popular branded products to consumers throughout China at a significant discount from retail prices. To optimize the shopping experience for their customers, the company built a personalized search recommendation system based on user query keywords and user portraits. Milvus, an open-source vector database built by Zilliz, is used to power the personalized recommender system on Vipshop’s mobile apps.

Objective

Build an e-commerce search recommendation system is to retrieve suitable products from a large number of products and display them to users according to their search intent and preference.

Challenges

  • Long computational response time - the average latency to retrieve TopK results from millions of items is around 300 ms.
  • High maintenance cost of ElasticSearch indexes - the same set of indexes is used for both commodity feature vectors and other related data, which hardly facilitates index building, but produces a massive amount of data.

Why Milvus

  • Support for various mainstream indexes makes vector search blazing fast.
  • Support for incremental and entire data update.
  • Support for distributed deployment, multi-language SDKs, and read/write separation.

Results

  • Searches on million-vector datasets complete in 30 milliseconds, approximately ten times faster than the previous solution adopted by Vipshop.
  • Provides accurate recommendations to users to optimize their shopping experience.

"At present, Milvus-based vector search can be used steadily in recommendation system scenarios, and its high performance gives us more room to play in the dimensionality of model and algorithm selection."

- Vipshop Search Service team