Recipe
publicShareStar

Build a rule-based chatbot from scratch

Collaborators
Reviewers
Total time needed: ~55 minutes
Learning Objectives
Understand how simple rule-based chatbots work and how to build one from scratch, understand their advantages and disadvantages in comparison to ML approaches.
Potential Use Cases
Build a chatbot application for personal or small simple business use cases.
Target Audience
BEGINNER
Go through the following annotated items in order:
LIBRARY 1. A refresher on regex
It helps you remember some simple regex so you can work out the codes of other tutorials better.
15 minutes
BOOK_CHAPTER 2. Explanation of types of Chatbots
This chapter will help you get an overview picture of what rule-based chatbots are, how they work and how they differ from the other more modern approaches.
10 minutes
BOOK_CHAPTER 3. Weizenbaum's explanation of ELIZA
An explanation of how the first rule-based chatbot which passed the Turing test shows how rules are created for the bot and inspire you on the power of chatbots.
20 minutes
REPO 4. Implementing ELIZA
Now it's time for you to implement an ELIZA for yourself and see how it works.
20 minutes
ARTICLE 5. A comparison between rule-based and ML chatbot
It explains well the situations in which you should choose a simple rule-based bot over a more complicated system and vice versa.
10 minutes
USE_CASE 6. Implement a simple rule-based chatbot
It helps you develop practical intuitions of what data is needed and the necessary steps to implement a chatbot.
10 minutes

Concepts Covered