Teach on Udemy

Turn what you know into an opportunity and reach millions around the world.

Learn More

Your cart is empty.

Keep shopping

Advanced WordPress Search - Full Text And Vector Search

Full Text Search, BM25, Vector/Semantic Search, Hybrid Search, Reranking and User Signals to enhance our search results

Created byCraig West

Last updated 2/2026

English

What you'll learn

Understand Full Text Search and its different search algorithms.
Learn what BM25 is and why it is so popular.
Learn Semantic Search and Vector Databases.
Learn how to perform Hybrid Search combining FTS and Semantic Search.
Learn how to rerank the results of hybrid search.

Course content

18 sections • 65 lectures • 4h 9m total length

Introduction8:15
eBook0:44
There is an eBook for reference in this course as well as the Markdown files used in the eBook available in the repo.
Repetition is the key to learning0:57
Repetition helps understanding. Many of these sections have overlap so one can skip parts as needed. The sections can be stand alone if needed.
The eBook files are also in your GitHub Repo0:06

Cline3:27
Cline is a VS Code extension that enables us to do AI Powered coding. It offers a number of free models which we can make use of.
Claude.ai2:52
How to use AI to get explanations of code3:29
Search is our focus not WordPress coding0:59
The course is about search not coding and plugin development. We have AI Assistance explaining the plugins in details and a way for you to ask AI questions on the code for fee. If you are looking to learn coding this is NOT the course for you.

Set up theme (optional) in a fresh install of WordPress3:19
We add our theme and child theme. This is optional as we will not be working in the front end and if we ever do, the theme is not important as we will be using plugins not theme files.
Add in the remaining plugins and load in posts with WP Run Importer5:26
We add in the remaining plugins and use WP Run Importer to load in the pre-made list of posts.
Create an OPENAI API Key1:13
For Vector Search we need embeddings. An LLM is needed and we set up our OpenAI Account and API Key in this lesson.
SQL Scripts Intro0:55
We load in our SQL scripts for our custom wp_products and articles tables.
Installing SQL Scripts1:27
We install the scripts for our custom tables wp_products and articles.

Term Frequency (TF)2:09
Inverse Document Frequency (IDF)4:00
Term Frequency - Inverse Document Frequency (TF-IDF)3:00
Best Match 25 (BM25)6:37
BM25 is the industry standard for key word search and is used in our MySQL FTS. We explore what it is in this lesson.
Negative BM25 Scores4:03
BM25 can give negative scores. This does not matter as we are looking for relative scores not absolute scores.
Stop Words2:55
Excessive indexes on a column0:28
We look at having redundant indexes on a column.
What WP MySQL FTS does3:19
We look at the overall process of what FTS does.

Requirements

Advanced developer knowledge of WordPress and MySQL
Access to an OpenAI key for vector search.

Description

This an advanced course for experienced WordPress Developers.

For Vector Search, and OpenAI API Key is needed. Currently, a minimum initial payment of $5USD is required. The course will use under $1USD with fair usage.

Full Text Search does not need an OpenAI API Key. You can follow the lectures without an API key.

Advanced WordPress Search in MySQL: FTS, BM25, Semantic Search, Hybrid Search, Reranking and Signals.

Modern websites live or die by the quality of their search experience. If your WordPress site relies on default search, you’re leaving relevance, speed, and user satisfaction on the table. This course gives you the technical depth and practical skills to build powerful, intelligent search systems.

What You’ll Learn

How WordPress search actually works under the hood, and why it struggles at scale.
Implementing MySQL Full‑Text Search (FTS) with boolean mode, natural language mode, and relevance tuning.
Using BM25 for dramatically improved ranking and scoring.
Building semantic search with embeddings and vector databases.
Applying reranking strategies to combine lexical and semantic signals - Hybrid Search.
Using Signals to assist rankings.

PLUGIN01 - A summary of TF, TF-IDF and BM25

PLUGIN02 - BM25 with adjustable `k1 ` and `b` parameters

PLUGIN03 - BM25 search results with calculations

PLUGIN04 - Tool yo compare TF, TF-IDF and BM25

PLUGIN05 - FTS on our custom wp_products table

PLUGIN20 - Building our FTS wp_posts_rag summary table with additional embeddings

PLUGIN21 - Searching our wp_posts_rag table

PLUGIN30 - Creating a set of endpoints for FTS, Vector and Hybrid search

PLUGIN31 - Reranking plugin for our set of FTS and Vector searches

PLUGIN32 - Creating a form for our Boolean FTS search

PLUGIN33 - Using our Boolean search to give a ranked set of results on our Posts

PLUGIN40 - A look at what Signals are and how they can help

Who this course is for:

Advance WordPress/MySQL users

Advanced WordPress Search - Full Text And Vector Search

What you'll learn

Explore related topics

Course content

Introduction4 lectures • 10min

My VSCode Setup3 lectures • 4min

AI Powered Coding Assistants I use - free and paid4 lectures • 11min

Restore final version of site with All in One Plugin1 lecture • 3min

Setting up our WordPress site from scratch5 lectures • 12min

TF, TF-IDF, BM25 and k1 and b parameters8 lectures • 27min

PLUGINS 01, 02, 03 and 04 on TF, TF-IDF and BM255 lectures • 11min

Implementations - MySQL Docs - 'articles' table7 lectures • 23min

Our custom wp_products table4 lectures • 25min

Our WP Posts4 lectures • 17min

Requirements

Description

Who this course is for: