Amazon OpenSearch Service now enables effortless integration with AI/ML models like Amazon SageMaker and Amazon Bedrock to power semantic search through a new neural search feature and AI/ML connectors.

<div>
<p>This article explains how to power semantic search using AI/ML connectors in Amazon OpenSearch Service. It demonstrates setting up a connector to an Amazon SageMaker model and using that model for semantic search on OpenSearch Service.</p>
<p>Specifically, the article covers:</p>
<ul>
<li>Overview of the neural search feature in OpenSearch Service 2.9</li>
<li>Prerequisites for setting up an AI/ML connector</li>
<li>Deploying a model on Amazon SageMaker using AWS CloudFormation</li>
<li>Verifying the deployed model on OpenSearch Dashboards</li>
<li>Creating a neural pipeline using the model ID</li>
<li>Creating a semantic search index using the neural pipeline</li>
<li>Ingesting sample data and validating vector generation</li>
<li>Comparing lexical and semantic search results</li>
<li>Cleaning up the deployed resources</li>
</ul>
</div>


Related articles