A Practical Guide to Vector Search Using Elasticsearch
 |
Description
Book Introduction
Learn how to optimize vector search, observability, cybersecurity, and ChatGPT integration with Elasticsearch!
While natural language processing (NLP) is primarily used for search use cases, this book aims to inspire others to leverage vectors to solve important domain challenges such as observability and cybersecurity.
Each chapter focuses on integrating vector search with Elasticsearch to improve not only search use cases but also observability and cybersecurity capabilities.
This book first introduces NLP and Elastic's capabilities in NLP processes.
Next, we'll look at how vectors are stored in a dense vector format, along with resource requirements and specific page cache requirements for fast response times.
As you continue reading, you'll discover various tuning techniques and strategies to improve your machine learning model deployment, including node scaling, configuration tuning, and load testing using Rally and Python.
We also cover vector search techniques using images, model fine-tuning for performance, and how to use the CLIP model for image similarity search in Elasticsearch.
Finally, we'll explore retrieval-augmented generation (RAG) and learn how to integrate ChatGPT with Elasticsearch to leverage vectorized data, the capabilities of ELSER, and the refined search mechanisms of RRF.
By the end of this book, you'll have all the skills you need to implement and optimize vector search in your projects using Elasticsearch.
While natural language processing (NLP) is primarily used for search use cases, this book aims to inspire others to leverage vectors to solve important domain challenges such as observability and cybersecurity.
Each chapter focuses on integrating vector search with Elasticsearch to improve not only search use cases but also observability and cybersecurity capabilities.
This book first introduces NLP and Elastic's capabilities in NLP processes.
Next, we'll look at how vectors are stored in a dense vector format, along with resource requirements and specific page cache requirements for fast response times.
As you continue reading, you'll discover various tuning techniques and strategies to improve your machine learning model deployment, including node scaling, configuration tuning, and load testing using Rally and Python.
We also cover vector search techniques using images, model fine-tuning for performance, and how to use the CLIP model for image similarity search in Elasticsearch.
Finally, we'll explore retrieval-augmented generation (RAG) and learn how to integrate ChatGPT with Elasticsearch to leverage vectorized data, the capabilities of ELSER, and the refined search mechanisms of RRF.
By the end of this book, you'll have all the skills you need to implement and optimize vector search in your projects using Elasticsearch.
- You can preview some of the book's contents.
Preview
index
[Part 1] The Basics of Vector Search
▣ 01: Introduction to Vectors and Embeddings
1.1 Examining the Role of Supervised and Unsupervised Learning in Vector Search
___What is an embedding/vector?
___What problems are vectors used to solve?
___Developer Environment
___Hugging Face
___Accelerating the market environment and developer experience
1.2 Use Cases and Applications
___AI-powered search
___Named Entity Recognition (NER)
___Sentiment Analysis
___Text classification
___Questions and Answers (QA)
___Text Summary
1.3 What role does Elastic play in this area?
___Basic Concepts of Observability and Cybersecurity
summation
▣ 02: Starting a Vector Search in Elastic
2.1 Vector Search Previous Search Experience in Elastic
___How Data Types Affect Relevance
___Relevance Model
2.2 Evolution of the search experience
Limitations of keyword-based search
___vector representation
2.3 New vector data type and vector search query API
___Sparse and dense vectors
___Getting Started with Elastic Cloud
___Dense Vector Mapping
___Brute-force kNN search
___kNN search
summation
[Part 2] Advanced Applications and Performance Optimization
▣ 03: Model Management and Vector Considerations in Elastic
3.1 Hugging Face
___Model Hub
___Datasets
___Spaces
3.2 Eland
Importing an embedding model from ___Hugging Face to Elasticsearch
___Configuring Elasticsearch Authentication
___Get models from Hugging Face Hub
Download the ___model
___Loading a model into Elasticsearch
Starting the ___ model
Deploying the ___model
___Create a query vector
3.3 Creating Vectors Inside Elasticsearch
3.4 Establishing Cluster Resource Planning
___CPU and memory requirements
___Disk Requirements
___Index Disk Usage Analysis API
3.5 Machine Learning Node Capacity
3.6 Storage Efficiency Strategies
___dimensional reduction
___quantization
Excluding dense_vector from ___source
summation
▣ 04: Performance Tuning - Verification through Data
4.1 Deploying the NLP Model
___Loading the model into Elasticsearch
___Settings related to model deployment
4.2 Load Testing
___Rally
___Memory (RAM) Usage Prediction
___Slowdown problem solved
summation
[Part 3] Special Use Cases
▣ 05: Image Search
5.1 Image Search Overview
___Image search advancements
___Image search method
___The role of vector similarity search
___Image search example
5.2 Image Vector Search
___Image Vectorization
___Indexing Image Vectors in Elasticsearch
___k-nearest neighbor (kNN) search
___Challenges and Limitations in Image Search
5.3 Multimodal Model for Vector Search
___The Need for Multimodality
___Understanding the vector space of multimodal models
___Introducing the OpenAI clip-ViT-B-32-multilingual-v1 model
___Applying vector search to various media types
summation
▣ 06: Removing Personally Identifiable Information Using Elasticsearch
6.1 PII and Redaction Overview
Data types that may include ___PII
Risks of PII stored in ___logs
___Types of PII leaks and losses
6.2 PII Removal Using NER Model and Regular Expression Patterns
___NER model
___regular expression pattern
Combining NER model and regular expression (Grok) patterns for ___PII removal
6.3 Elasticsearch's PII Removal Pipeline
___Creating fake PII
___Basic pipeline settings
___Expected Results
6.4 Options for Extending and Detailing the PII Removal Pipeline
___Customizing the Basic PII Example
___Duplicate the pipeline and create a new version for the new data stream.
___Fine-tuning a NER model for a specific dataset
___The need for context-aware technology and how to apply it
summation
▣ 07: Vector-based next-generation observability
7.1 Introduction to Observability and Its Importance in Modern Software Systems
___Observability - Key Elements
___Log Analysis and Its Role in Observability
7.2 A New Approach to Applying Vectors and Embeddings to Log Analysis
___Approach 1 - Train or fine-tune an existing model for logs
___Approach 2 - Generate human-readable descriptions, then vectorize them.
7.3 Log vectorization
___synthetic log
___Log Expansion with OpenAI
7.4 Log Semantic Search
___log vector index
___Model Loading
___Collection Pipeline
___Semantic Search
summation
▣ 08: The Impact of Vectors and Embeddings on Strengthening Cybersecurity
8.1 Understanding the Importance of Email Phishing Detection
___What is phishing?
___Various types of phishing attacks
___Statistics on the frequency of phishing attacks
___The Challenges of Phishing Email Detection
___The role of automatic detection
___Supplementing existing technologies with natural language processing technology
8.2 Introduction to ELSER
8.3 The Role of ELSER in Generative AI
8.4 Enron Email Data Set (Ham or Spam)
8.5 Seeing ELSER in action
___Hardware Considerations
___Download the ELSER model to Elasticsearch
___Index setup and data collection pipeline settings
Semantic search using ___ELSER
Limitations of ___ELSER
summation
[Part 4] Innovative Integration and Future Direction
▣ 09: Augmented Search Creation with Elastic
9.1 Preparing for RAG Enhanced Search Using ELSER and RRF
Semantic search using ___ELSER
Summary of Essential Considerations for ___RAG
ELSER Integration Using ___RRF
___Language Model and RAG
9.2 In-Depth Case Study - Building a RAG-Based CookBot
___Dataset Overview - Explore the Allrecipes.com Dataset
___Preparing data for RAG enhanced search
RRF finder using ___ELSER
___Using the search engine and creating a generator
summation
▣ 10: Building an Elastic Plugin for ChatGPT
10.1 Contextual Basics
___Dynamic Context Paradigm
10.2 DCL Plugin - Structure and Operation
10.3 Implementing DCL
___Get the latest information from the Elastic documentation
___Raising the Data Level with Embedchain
___Integrate with ChatGPT to create real-time conversation partners
___distribution
summation
▣ Appendix: Creating Elastic Guide GPT
▣ 01: Introduction to Vectors and Embeddings
1.1 Examining the Role of Supervised and Unsupervised Learning in Vector Search
___What is an embedding/vector?
___What problems are vectors used to solve?
___Developer Environment
___Hugging Face
___Accelerating the market environment and developer experience
1.2 Use Cases and Applications
___AI-powered search
___Named Entity Recognition (NER)
___Sentiment Analysis
___Text classification
___Questions and Answers (QA)
___Text Summary
1.3 What role does Elastic play in this area?
___Basic Concepts of Observability and Cybersecurity
summation
▣ 02: Starting a Vector Search in Elastic
2.1 Vector Search Previous Search Experience in Elastic
___How Data Types Affect Relevance
___Relevance Model
2.2 Evolution of the search experience
Limitations of keyword-based search
___vector representation
2.3 New vector data type and vector search query API
___Sparse and dense vectors
___Getting Started with Elastic Cloud
___Dense Vector Mapping
___Brute-force kNN search
___kNN search
summation
[Part 2] Advanced Applications and Performance Optimization
▣ 03: Model Management and Vector Considerations in Elastic
3.1 Hugging Face
___Model Hub
___Datasets
___Spaces
3.2 Eland
Importing an embedding model from ___Hugging Face to Elasticsearch
___Configuring Elasticsearch Authentication
___Get models from Hugging Face Hub
Download the ___model
___Loading a model into Elasticsearch
Starting the ___ model
Deploying the ___model
___Create a query vector
3.3 Creating Vectors Inside Elasticsearch
3.4 Establishing Cluster Resource Planning
___CPU and memory requirements
___Disk Requirements
___Index Disk Usage Analysis API
3.5 Machine Learning Node Capacity
3.6 Storage Efficiency Strategies
___dimensional reduction
___quantization
Excluding dense_vector from ___source
summation
▣ 04: Performance Tuning - Verification through Data
4.1 Deploying the NLP Model
___Loading the model into Elasticsearch
___Settings related to model deployment
4.2 Load Testing
___Rally
___Memory (RAM) Usage Prediction
___Slowdown problem solved
summation
[Part 3] Special Use Cases
▣ 05: Image Search
5.1 Image Search Overview
___Image search advancements
___Image search method
___The role of vector similarity search
___Image search example
5.2 Image Vector Search
___Image Vectorization
___Indexing Image Vectors in Elasticsearch
___k-nearest neighbor (kNN) search
___Challenges and Limitations in Image Search
5.3 Multimodal Model for Vector Search
___The Need for Multimodality
___Understanding the vector space of multimodal models
___Introducing the OpenAI clip-ViT-B-32-multilingual-v1 model
___Applying vector search to various media types
summation
▣ 06: Removing Personally Identifiable Information Using Elasticsearch
6.1 PII and Redaction Overview
Data types that may include ___PII
Risks of PII stored in ___logs
___Types of PII leaks and losses
6.2 PII Removal Using NER Model and Regular Expression Patterns
___NER model
___regular expression pattern
Combining NER model and regular expression (Grok) patterns for ___PII removal
6.3 Elasticsearch's PII Removal Pipeline
___Creating fake PII
___Basic pipeline settings
___Expected Results
6.4 Options for Extending and Detailing the PII Removal Pipeline
___Customizing the Basic PII Example
___Duplicate the pipeline and create a new version for the new data stream.
___Fine-tuning a NER model for a specific dataset
___The need for context-aware technology and how to apply it
summation
▣ 07: Vector-based next-generation observability
7.1 Introduction to Observability and Its Importance in Modern Software Systems
___Observability - Key Elements
___Log Analysis and Its Role in Observability
7.2 A New Approach to Applying Vectors and Embeddings to Log Analysis
___Approach 1 - Train or fine-tune an existing model for logs
___Approach 2 - Generate human-readable descriptions, then vectorize them.
7.3 Log vectorization
___synthetic log
___Log Expansion with OpenAI
7.4 Log Semantic Search
___log vector index
___Model Loading
___Collection Pipeline
___Semantic Search
summation
▣ 08: The Impact of Vectors and Embeddings on Strengthening Cybersecurity
8.1 Understanding the Importance of Email Phishing Detection
___What is phishing?
___Various types of phishing attacks
___Statistics on the frequency of phishing attacks
___The Challenges of Phishing Email Detection
___The role of automatic detection
___Supplementing existing technologies with natural language processing technology
8.2 Introduction to ELSER
8.3 The Role of ELSER in Generative AI
8.4 Enron Email Data Set (Ham or Spam)
8.5 Seeing ELSER in action
___Hardware Considerations
___Download the ELSER model to Elasticsearch
___Index setup and data collection pipeline settings
Semantic search using ___ELSER
Limitations of ___ELSER
summation
[Part 4] Innovative Integration and Future Direction
▣ 09: Augmented Search Creation with Elastic
9.1 Preparing for RAG Enhanced Search Using ELSER and RRF
Semantic search using ___ELSER
Summary of Essential Considerations for ___RAG
ELSER Integration Using ___RRF
___Language Model and RAG
9.2 In-Depth Case Study - Building a RAG-Based CookBot
___Dataset Overview - Explore the Allrecipes.com Dataset
___Preparing data for RAG enhanced search
RRF finder using ___ELSER
___Using the search engine and creating a generator
summation
▣ 10: Building an Elastic Plugin for ChatGPT
10.1 Contextual Basics
___Dynamic Context Paradigm
10.2 DCL Plugin - Structure and Operation
10.3 Implementing DCL
___Get the latest information from the Elastic documentation
___Raising the Data Level with Embedchain
___Integrate with ChatGPT to create real-time conversation partners
___distribution
summation
▣ Appendix: Creating Elastic Guide GPT
Detailed image
Publisher's Review
★ What this book covers ★
◎ Optimizing performance using vector search functions
◎ Exploring image vector search and its applications
◎ Detecting and masking personally identifiable information
◎ Implementing log analysis and search for next-generation observability
◎ Utilizing vector-based bot detection for cybersecurity
◎ Explore vector space visualization and Elasticsearch's latest search capabilities.
◎ Implementing a RAG-enhanced application using Streamlit
◎ Optimizing performance using vector search functions
◎ Exploring image vector search and its applications
◎ Detecting and masking personally identifiable information
◎ Implementing log analysis and search for next-generation observability
◎ Utilizing vector-based bot detection for cybersecurity
◎ Explore vector space visualization and Elasticsearch's latest search capabilities.
◎ Implementing a RAG-enhanced application using Streamlit
GOODS SPECIFICS
- Date of issue: June 26, 2024
- Page count, weight, size: 276 pages | 175*235*14mm
- ISBN13: 9791158395223
You may also like
카테고리
korean
korean
![GQ KOREA Mark (Monthly): December [2025]](http://librairie.coreenne.fr/cdn/shop/files/8ef265dbbfbf186523ed75ba7319009d.jpg?v=1765340328&width=3840)
