ChatterBot is a machine-learning based conversational dialog engine build in Python which makes it possible to generate responses based on collections of known conversations. The language independent design of ChatterBot allows it to be trained to speak any language.

An example of typical input would be something like this:

user: Good morning! How are you doing?
bot: I am doing very well, thank you for asking.
user: You're welcome.
bot: Do you like hats?

An untrained instance of ChatterBot starts off with no knowledge of how to communicate. Each time a user enters a statement, the library saves the text that they entered and the text that the statement was in response to. As ChatterBot receives more input the number of responses that it can reply and the accuracy of each response in relation to the input statement increase. The program selects the closest matching response by searching for the closest matching known statement that matches the input, it then returns the most likely response to that statement based on how frequently each response is issued by the people the bot communicates with.

This package can be installed from PyPi by running:

pip install chatterbot

from chatterbot import ChatBot
chatbot = ChatBot("Ron Obvious")

Note: *The ChatBot requires that a name is specified for the bot.

After creating a new chatterbot instance it is also possible to train the bot. Training is a good way to ensure that the bot starts off with knowledge about specific responses. The current training method takes a list of statements that represent a conversation.

Note: Training is not required but it is recommended.

conversation = [
    "Hello",
    "Hi there!",
    "How are you doing?",
    "I'm doing great.",
    "That is good to hear",
    "Thank you.",
    "You're welcome."
]

chatbot.train(conversation)

response = chatbot.get_response("Good morning!")
print(response)

Your ChatterBot will learn based on each new input statement it recieves. If you do not want your bot to learn, set logging=False when initializing the bot.

chatbot = ChatBot("Johnny Five", logging=False)

ChatterBot uses adapters to handle three types of operations. All adapters for ChatterBot fall into one of three categories: storage, io, and logic.

By default, ChatterBot uses the JsonDatabaseAdapter adapter for storage, the ClosestMatchAdapter for logic, and the TerminalAdapter for IO.

Each adapter can be set by passing in the dot-notated import path to the constructor.

bot = ChatBot("My ChatterBot",
    storage_adapter="chatterbot.adapters.storage.JsonDatabaseAdapter",
    logic_adapter="chatterbot.adapters.logic.ClosestMatchAdapter",
    io_adapter="chatterbot.adapters.io.TerminalAdapter",
    database="../database.db")

Storage adapters allow ChatterBot to connect to connect to any type of storage backend.

"chatterbot.adapters.storage.JsonDatabaseAdapter"

The JSON Database adapter requires an additional parameter (database) to be passed to the ChatterBot constructor. This storage adapter uses a local file database so this parameter is needed to specify the location of the file.

IO adapters allow ChatterBot to communicate through various interfaces. The default io adapter uses the terminal to communicate with the user.

"chatterbot.adapters.io.TerminalAdapter"

The terminal adapter allows the ChatterBot to communicate with you through your terminal.

Logic adapters determine how ChatterBot selects responces to input statements.

"chatterbot.adapters.logic.ClosestMatchAdapter"

The ClosestMatchAdapter selects a response based on the closest know match to a given statement.

For examples, see the examples directory in this project's repository.

ChatterBot's built in tests can be run using nose.
See the nose documentation for more information.

Using ChatterBot in your app? Let us know!

See release notes for changes https://github.com/gunthercox/ChatterBot/releases