Build a Python Chatbot with ChatterBot: Step‑by‑Step Guide

Preface

In the age of artificial intelligence, chatbots have become increasingly popular as the latest tool for simplifying human‑computer interaction, from e‑commerce to healthcare.

What is a Chatbot

A chatbot is a software application that conducts online conversations via text or text‑to‑speech instead of direct contact with a human agent (according to Wikipedia).

Types of Chatbots

Chatbots can be divided into two categories:

Rule‑based

Self‑learning

Rule‑based: These bots answer questions based on pre‑defined rules and are suitable for simple queries.

Self‑learning chatbots: Powered by machine‑learning algorithms, they can learn from data and become smarter over time.

How Chatbots Work

AI‑driven chatbots use natural language processing and machine‑learning algorithms to learn from data, understand user input, and generate responses in text or speech. Examples include Google Assistant, Alexa, and Siri.

Because they can learn from behavior and experience, they can handle a wide range of queries and commands.

Getting Started

We will create a Python chatbot using the ChatterBot library.

1. Create a Virtual Environment

pipenv is a convenient Python tool for creating virtual environments.

pip install pipenv
pipenv install

2. Install Libraries

Install ChatterBot and its corpus data.

pipenv install chatterbot
pipenv install chatterbot_corpus

3. Create and Train the Chatbot

from chatterbot import ChatBot
from chatterbot.trainers import ChatterBotCorpusTrainer
BOTNAME = "Pyter"

def start():
    bot = ChatBot(
        BOTNAME,
        logic_adapters=[
            {
                'import_path': 'chatterbot.logic.BestMatch',
                'default_response': 'I am sorry, but I do not understand.',
                'maximum_similarity_threshold': 0.90,
            },
        ],
        preprocessors=["chatterbot.preprocessors.clean_whitespace"],
        input_adaptor="chatterbot.input.TerminalAdaptor",
        output_adaptor="chatterbot.output.TerminalAdaptor",
        database_uri='sqlite:///database.sqlite3'
    )
    trainer = ChatterBotCorpusTrainer(bot)
    # Train based on the English corpus
    trainer.train(
        "chatterbot.corpus.english",
        "chatterbot.corpus.english.greetings",
        "chatterbot.corpus.english.conversations",
    )
    print(f"Hello I am {BOTNAME}")
    while True:
        try:
            bot_input = input("You: ")
            bot_respose = bot.get_response(bot_input)
            print(f"{BOTNAME}: {bot_respose}")
        except (KeyboardInterrupt, EOFError, SystemExit):
            break

if __name__ == "__main__":
    start()

Running the script launches an interactive terminal chatbot that responds to user input based on the trained data.