Elasticsearch meets BERT: Building Search Engine with Elasticsearch and BERT

In this post, we use a pre-trained BERT model and Elasticsearch to build a search engine. Elasticsearch has recently released text similarity search with vector fields. On the other hand, you can convert text into a fixed-length vector using BERT. So once we convert documents into vectors by BERT and store them into Elasticsearch, we can search similar documents with Elasticsearch and BERT.

This post implements a search engine with Elasticsearch and BERT by the following architecture. Here, we use…