NativeChat Documentation

Learn how to develop, train, test and publish chatbots with NativeChat.

Getting Started

Read through some general explanations about what NativeChat platform can do and how it works.

Overview - Introduction to the basic terms of chatbot development and development workflow of NativeChat.
Design & Planning - A list of best practices when developing a chatbot.
Tutorial: Comprehensive - In-depth guide on all important NativeChat concepts.
Tutorial: Quickstart - Shows you how to build a sample chatbot for booking an appointment with a doctor.

Learn how to define conversations and steps in each conversation of your bot.

Overview - Explains the basic terms used when defining a cognitive flow.
Conversations - Explains the different types of conversations.
Steps Configuration - Documentation for the configuration of all types of steps and their properties.
Working with Bot Memory - How to store, access and format data from bot memory.
Settings - Configuration of the general bot settings.
Authentication - Configuration of 3rd-party authentication from the bot.

Learn how to train your bot’s understanding of what users say.

Blog post: What is Natural Language Processing (NLP)?
Conversation triggers - Help your chatbot identify the correct conversation from the Cognitive Flow that needs to be started.
Entities - Explains the basic terms used when defining a cognitive flow.
Question answering - Train your bot for Small Talk and Frequently Asked Questions.
Dynamic Training - Configure NativeChat to use training data from your system by executing an REST API call.
Import Training Definition - Bulk-import training data using JSON definition.
Understanding assessment - Annotate user phrases to enrich FAQ capabilities.
Test set - See and modify phrases added through Understanding assessment.

Tips and tricks for testing and debugging your bot.

Overview - Introduction to testing and debugging NativeChat bots.
Blog post: 5 Handy Tricks for Improving Your Chatbot Understanding with NativeChat - Introduction to testing and debugging NativeChat bots.
Developer Productivity - How to test & debug your chatbot on various channels.
Training Problems Check - Analyze your NLP training data for problems.
Proxy bots - Use NativeChat’s proxy bots for Facebook Messenger and Amazon Alexa to test on these platforms before going live.

NativeChat bot can work on variety of channels - web, mobile, or social media. Check the how-to guides on publishing yours.

Learn more about the analytics data provided by NativeChat. Use this information to improve your bot’s performance and functionality over time.

Overview - Introduction to the metrics.
Engagement Metrics - Learn how many people are using your bot and how successful they are.
User Retention Metrics - Learn how many unique users interact with your bot and how much time they spend chatting with it.
Understanding Metrics - Learn how well your bot understands what the users say.
User Feedback Metrics - Learn how positive your users rate the usefulness of their interaction with the bot.
Search Metrics - Learn how often your bot offers search for not recognized phrases, how often users accept and how positive they rate the search results.
Channel Metrics - Learn which channels you have published your bot on are preferred by your users.

Collaboration Access - Give administartive access to your bot to collaborators.
Authentication Methods - Advanced authentication for endpoints used in the webhook steps of your bot.
External History Storage - Configure your bot to send each completed conversation’s data to the specified endpoint and delete it from Progress’ data storage systems
Low Confidence Suggestions - Configure your bot to provide suggestions when it understands the user message with low confidence.
Search Fallback - Configure your bot to search for an expression when it does not understand the user message.
User Feedback - Configure your bot to ask the user for feedback after it has answered an FAQ, finished a conversation or executed search.
Live Agent Hand-off - Configure live agents to which your bot can transfer the session when instructed by the user.