Natural Language Understanding

NativeChat's built-in Natural Language Processing engine allows developers to extract structured data and user intent from free text and image replies.

The Challenge

Training a bot to converse about existing business entities often requires duplicating data and lengthy training processes.

The Solution

NativeChat supports dynamic training on top of existing enterprise data through web services from Salesforce, Workday, Zendesk or any other enterprise product.

Get Started

Turn Free Text Into Structured Data

By providing just a few examples of how people talk about the task they want to accomplish, the bot can start to identify user intent and respond accordingly.

The Natural Language Understanding also helps you extract the usable information bits (or entities) from free text replies.

Leverage Existing Data With Dynamic Training

NativeChat bots can be trained to recognize entities based on data already available in your systems. You just point it to a standard REST API endpoint and it handles the training from there—and keeps it up-to-date.

For example, you can train a bot to understand the names of hospital personnel when users are asking about a specific individual, or identify a contact from your CRM system based on some of its properties.

Change of Intent Mid-Conversation

Because every user input is parsed by the NLP engine, the chatbot can detect when a user wants to change the topic of a conversation and respond appropriately.

FAQ and Small Talk

In real conversations, people often interject a question because they've just thought of it.

For example, when you're booking an appointment for the first time in a hospital, it's normal to ask "Okay, but how much would it cost?"

NativeChat handles these "interruptions" by providing the appropriate answer and then returns to the main conversation topic, trying to achieve the initial goal of the conversation.