Simple Chatbot using NLTK

  Simple Chatbot using NLTK

NLTK stands for Natural language toolkit used to deal with NLP applications. A chatbot is one among those applications. This post explains Rule-based chatbots using the NLTK library. 

Terms you should know.

Chat – Chat is a class that contains complete logic for processing the text data which the chatbot receives.

Reflections – Reflections is a dictionary containing basic input and corresponding outputs. You can create your own dictionary with more responses you want. 

Steps:

1)  Please install the NLTK library first before working using the pip command.

pip install nltk

2)  Next is to import the library and classes we need to use.

import nltk

from nltk.chat.util import Chat, reflections

3) After importing the libraries, we have to create rules. 

The lines of code given below create a simple set of rules. The first line describes the user input which we have taken as raw string input and the next line is our chatbot response. You can modify these pairs as per the questions and answers you want.  The nltk.chat works on various regex patterns present in user Intent and corresponding to it, presents the output to a user. 

 Here is the Code:

import nltk

from nltk.chat.util import Chat, reflections

 pairs = [

    [

        r"my name is (.*)",

        ["Hello %1, How are you today ?",]  

    ],

    [

        r"hi|hey|hello",

        ["Hello", "Hey there","Namskar"]  # it outputs one of this response randomly

    ],

    [

        r"what is your name ?",

        ["I am a bot created to assist you. You can call me Rupali-Bot!",]

    ],

    [

        r"how are you ?",

        ["I'm Super fine ! Hope are too",]

    ],

    [

        r"sorry (.*)",

        ["Its alright", "Never mind", "It happens, don't mind"]

    ],

    [

        r"I am fine",

        ["Great to hear that, How can I help you?",]

    ],

       [

        r"what (.*) want ?",

        ["Make me an offer I can't refuse",]

    ],

    [

        r"(.*) created ?",

        ["Rupali created me using Python's NLTK library ","It is a secret",]

    ],

    [

        r"(.*) (location|city) ?",

        ['Nasik, Maharshtra',]

    ],

    [

        r"how is weather in (.*)?",

        ["Weather in %1 is awesome like always","%1 has always fantastic weather ","Not Too cold not too hot in %1",]

    ],

    [

        r"i work in (.*)?",

        ["%1 is an Amazing company, I have heard about it. But make sure to get more details.",]

    ],

   

    [

        r"how (.*) health(.*)",

        ["I'm a computer program, so I'm always at best of my health ", "Super Fine",]

    ],

    [

        r"(.*) (sports|game) ?",

        ["I'm a very big fan of IPL thats Cricket",]

    ],

    [

        r"who (.*) sportsperson ?",

        ["Virat Kohli","Sachin Tendulakr","M.S. Dhoni"]

    ],

    [

        r"who (.*) (moviestar|actor)?",

        ["Aamir Khan", "Pariniti Chopra",]

    ],

 

    [

        r"quit",

        ["BBye take care. See you soon :) ","It was nice talking to you. See you soon :)"]

    ],

]

 

def chat():

    print("Hi! I am a chatbot created by Rupali for your assistance")

    chat = Chat(pairs, reflections)

    chat.converse()

#initiate the conversation

if __name__ == "__main__":

    chat()

 

-------------------------------------- The output---------------------------------------

Will you try and see it ? 

Ok..here is it !!

Today most Chatbots are created using tools like Dialogflow, RASA, etc.