Implementing semantic search with dense vectors

In this recipe, we will explore the fundamentals of vector search. We’ll demonstrate how to implement semantic search for the movie Search Application we built in Chapter 3. Elasticsearch supports both dense and sparse vectors – these are the two primary types of data representations commonly used to convert text or other types of data into a numeric form known as vectors. For this recipe, our focus will be on dense vectors. In the next recipe, we will delve into sparse vectors and examine the key characteristics of each type.

Getting ready

We will reuse some concepts that we learned in Chapter 3 and Chapter 8:

• Ensure that you are familiar with search templates and search applications
• Ensure that you understand the concept of importing third-party machine learning models
• The snippets for this recipe are available at https://github.com/PacktPublishing/Elastic-Stack-8.x-Cookbook/blob/main/Chapter9/snippets...