Chatbot Basics for Customer Communication
Download as PPTX, PDF1 like169 views
This document provides a beginner's guide to chatbots, explaining the differences between rule-based and AI-based chatbots. It highlights how rule-based chatbots work through predefined responses, while AI chatbots leverage natural language processing to understand context and user intent. Additionally, the document discusses platforms for creating chatbots and offers a brief guide on setting up a chatbot for customer communication using the Kompose platform.
Chatbot Basics for Customer Communication
- 1. Chatbot Basics for Customer Communication Beginner’s Guide to Chatbots
- 2. What is a Chatbot? ● Computer program designed to simulate human conversation ● Interpret the words given to them and provide a pre-set answer Example
- 3. Types of Chatbots Rule based chatbots AI based chatbots
- 4. Rule based chatbots ● User input is matched using predefined rules in order to get an answer ● Mostly used as button based bots
- 5. How Rule Based Chatbot works ● Prepare list of user inputs, responses and the rules ● User input is matched using predefined rules in order to get an answer User says Rule Bot replies hey Exact match Hey there nice to talk to you Exact match I enjoy talking to you, too. bye Contains Bye, have a good day!
- 6. Rule based chatbots code sample ● Following pseudo code shows a simplest view of how it works. Example if user input matches ‘hey’ then reply ‘hey there’ if user input contains ‘bye’ then reply ‘Bye, have a good day!’
- 7. Limitations of Rule Based Chatbots Result accuracy is not good Hard to maintain for large data set Fails in complex queries
- 8. AI based chatbots ● Not just words, AI chatbot understands what someone is saying ● AI chatbots understand context and the intent of the user, then generates the answer
- 9. How does AI Chatbot Works? ● AI Chatbot uses Natural Language Processing (NLP) to break down the user sentence into: ○ Intents: Intents are action ○ Entity: Details that compliment the intent ● Intents and Entity information along with Context is used to generate the response
- 10. Natural Language Processing (NLP) ● Natural Language refers to the way humans communicate with each other. ● Natural Language Processing (NLP) is broadly defined as the electronic manipulation of natural language, like speech and text, by software. ● NLP involves the reading and understanding of human’s spoken and written language through the computer
- 11. Intents ● Intent represents the purpose of user input ● If a user says ‘I want to book a demo’, then the user’s intention is to ‘Book demo’. Phrases Intent Hi Hey Hello Greeting What is the price? What is the cost? How much does it cost? Cost
- 12. Entity ● Entity represents details that compliment the intent Phrases Entities When is Tesla coming to India? Tesla, India What is price for Model S? Model S Book a demo for tomorrow tomorrow
- 13. Top Chatbots platforms to use Dialogflow IBM Watson RASA Kompose by Kommunicate Amazon Lex Learn more about differences between above platforms here.
- 14. Create Chatbot for customer communication ● Let’s go through a sample on how to create a chatbot. ● We will be using Kompose NLP based chatbot platform for the example as it provides all essentials things for customer communication. Kompose is built specifically for the customer communication. ● Before we proceed, signup at Kommunicate.io and create a bot from Kompose section.
- 15. Chatbot Setup We will setup 3 things in chatbot Welcome Message Human HandoverIntents
- 16. Welcome Message ● Welcome message is the first message which is displayed to users when they first visit the chatbot ● A good welcome message is important to help the user understand what the chatbot is all about Example
- 18. Intents ● Kommunicate call intents as “Answers” ● Add training phrases and the answer Example
- 19. Human handover ● For personalized support and complex queries, it is advisable to handover chat to the human agent. ● Kommunicate provides an easy setting at the time of bot creation for bot to human handoff. Example
- 21. Thank You! Devashish Datt Mamgain Co-founder & CEO devashish@kommunicate.io www.kommunicate.io