Skip to content

Commit

Permalink
Updates based on feedback
Browse files Browse the repository at this point in the history
  • Loading branch information
kosabogi committed Nov 20, 2024
1 parent f465344 commit 68ce33e
Showing 1 changed file with 16 additions and 15 deletions.
31 changes: 16 additions & 15 deletions serverless/pages/search-your-data-semantic-search.asciidoc
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,12 @@ Semantic search is a search method that helps you find data based on the intent

Elasticsearch provides various semantic search capabilities using natural language processing (NLP) and vector search. Using an NLP model enables you to extract text embeddings out of text. Embeddings are vectors that provide a numeric representation of a text. Pieces of content with similar meaning have similar representations.

There are three main workflows for implementing semantic search with {es}, arranged in order of increasing complexity:

- <<elasticsearch-reference-semantic-search-semantic-text,The `semantic text` workflow>>
- <<elasticsearch-reference-semantic-search-inference-api,Inference API workflow>>
- https://www.elastic.co/guide/en/elasticsearch/reference/current/semantic-search-deployed-nlp-model.html[Model deployment workflow]

image::images/semantic-options.svg[Overview of semantic search workflows in {es}]

[NOTE]
Expand All @@ -18,23 +24,21 @@ Semantic search is available on all Elastic deployment types: self-managed clust
====

[discrete]
[[lasticsearch-reference-semantic-search-inference-api]]
== Semantic search with the inference API
[[elasticsearch-reference-semantic-search-semantic-text]]
== Semantic search with `semantic text`

The inference API workflow enables you to perform semantic search using models from a variety of services, such as Cohere, OpenAI, HuggingFace, Azure AI Studio, and more.
The `semantic_text` field simplifies semantic search by providing inference at ingestion time with sensible default values, eliminating the need for complex configurations.

Learn how to implement semantic search with the inference API in the {ref}/semantic-search-inference.html[Elasticsearch docs →].
Learn how to implement semantic search with `semantic text` in the https://www.elastic.co/guide/en/elasticsearch/reference/current/semantic-search-semantic-text.html[Elasticsearch docs →].

[discrete]
[[elasticsearch-reference-semantic-search-semantic-text]]
== Semantic search with `semantic text`
[[elasticsearch-reference-semantic-search-inference-api]]
== Semantic search with the inference API

The `semantic_text` field simplifies semantic search by providing inference at ingestion time with sensible default values, eliminating the need for complex configurations.
The inference API workflow enables you to perform semantic search using models from a variety of services, such as Cohere, OpenAI, HuggingFace, Azure AI Studio, and more.

Learn how to implement semantic search with the inference API in the {ref}/semantic-search-inference.html[Elasticsearch docs →].

[NOTE]
====
ℹ️ The Semantic search with `semantic text` documentation currently lives in the https://www.elastic.co/guide/en/elasticsearch/reference/current/semantic-search-semantic-text.html[Elasticsearch docs].
====

[discrete]
[[elasticsearch-reference-semantic-search-elser]]
Expand All @@ -46,7 +50,4 @@ representation. Instead of literal matching on search terms, semantic search
retrieves results based on the intent and the contextual meaning of a search
query.

[NOTE]
====
ℹ️ The Semantic search with ELSER documentation currently lives in the https://www.elastic.co/guide/en/elasticsearch/reference/current/semantic-search-elser.html[Elasticsearch docs].
====
Learn how to implement semantic search with ELSER in the https://www.elastic.co/guide/en/elasticsearch/reference/current/semantic-search-elser.html[Elasticsearch docs →].

0 comments on commit 68ce33e

Please sign in to comment.