Elastic Stack in Action: Discover the Power of Semantic Search

Overview

Semantic search is transforming how users discover information—moving beyond exact keyword matches to understanding intent and meaning. In this hands-on training workshop, participants will learn how to build a complete semantic search pipeline using Elasticsearch, Kibana, and the ELSER model.

You will configure an inference endpoint, create a semantic-aware index using the semantic_text field, ingest a real-world movie dataset, and run both traditional keyword searches and modern semantic searches to clearly see the difference in relevance and user experience.

This workshop is designed to be practical, interactive, and production-focused, giving you the skills needed to bring semantic search into real applications.

Learning Goals

• By the end of this training, participants will be able to:
• Understand the fundamentals of semantic search vs traditional keyword search
• Configure an inference endpoint for the ELSER semantic model
• Create Elasticsearch index mappings using the semantic_text field type
• Ingest and index a real movies dataset for semantic retrieval

• Execute and compare:

•  Keyword-based match queries
•  Semantic queries for meaning-based search

• Identify real-world use cases where semantic and hybrid search significantly improve relevance
• Understand how semantic search fits into modern search-driven applications

Agenda

1. Introduction & Foundations

• What is semantic search and why it matters
• Limitations of keyword-based search
• Overview of semantic search architecture in Elasticsearch

2. Core Concepts Deep Dive

• What is the semantic_text field
• What is ELSER and how it works
• Inference endpoints and embedding generation
• Understanding semantic vs keyword queries

3. Hands-on Lab: Data Ingestion, Search & Comparison

• Configuring the ELSER inference endpoint
• Creating a semantic-enabled index mapping
• Ingesting the movies dataset
• Validating indexed semantic fields
• Verifying embedding generation
• Running traditional match queries
• Performing semantic search on movie plots
• Comparing relevance and ranking behavior
• Introduction to hybrid search concepts

4. Wrap-up, Use Cases & Q&A 

• When to use keyword, semantic, or hybrid search
• Real-world implementation considerations
• Open Q&A and discussion