Tokopedia’s AI-enabled E-commerce Site Search

Tokopedia is a technology company specializing in e-commerce. Its marketplace is the most popular online shopping platform in Indonesia, with 100+ million monthly active users and 9.7+ million merchants. To make search results more relevant and improve user engagement, Tokopedia built a smart search engine capable of pairing low fill rate keywords with high fill rate keywords. Milvus, an open-source vector database built by Zilliz, is used to index the underlying product database and optimize Tokopedia's AI-enabled shopping experience.
Objective
Build a search engine capable of matching low fill rate keywords with high fill rate keywords to return relevant search results from massive datasets instantaneously.
Challenges
- Return accurate search results in milliseconds from massive vector datasets. - Search functionality must remain highly available amid a node, or system wide, breakdown.
Why verctor database
- Support for various mainstream indexes makes vector search blazing fast. - Mishards, Milvus' cluster sharding middleware, makes vector search highly available. - Milvus-Helm simplifies and accelerates node configurations.
Results
Relevance of search results and CVR are improved significantly.
"We chose Milvus because it is very user-friendly. All we have to do is simply update the parameters based on our application scenario. It has detailed documentation and, if you run into issues, you can always find support from its community. Without Milvus, our semantic search wouldn’t have been 10x smarter!"
Read the full storyshare this
