Photo by Felix Mittermeier on Unsplash

AI & CODING

🤖 How to talk to Computers: A Framework for building Conversational Agents — Part 2

Theory and Practice with a sample project

This post is the 2nd part in the multipart series on building a conversational agent with a Chatbot and Recommendation System integration, and build with React, Python, and Node.js 🌌

In the 1st part, you learned:

• A brief summary of Chatbots and Recommendation Systems and how we will be implementing them ✏️
• How to train a Chatbot on any set of rules with Natural Language Processing on Wit.ai 💬
• A method for implementing a simple Recommendation Engine from Scratch with Flask & Python 🐍

Here is the link to the 1st part:

🤖 How to Talk to Computers: A Framework for building Conversational Agents — Part 1

And here is a link to part 3:

🤖 How to talk to Computers: A Framework for building Conversational Agents — Part 3

And here is the demo of the final product we are building and from which we will be learning 📕

Loa🧙‍♀️ — A small AI-based conversational agent capable of providing accurate information about U.S located restaurants, and integrated with a content-based recommendation engine.

What you will learn

Here is an updated version of what you will learn in this part:

• How to connect our Chatbot and Recommendation Services together with a global API in NodeJS. 🏗️
• How to bootstrap a REST API serving our systems from scratch, that will be consumed in the front-end ⚛️
• A light overview of deployment & integration with Docker, Heroku and Netlify 📦

More specifically, if we look at a quick diagram of our project architecture, this is what we will be focusing on today: