In order to create Dialog Tasks, the Dialog Builder uses nodes and transitions to make the necessary connections between the components of your conversation. In this article, we will overview the available node types, transitions, as well as context object information.
How do Nodes and Transitions Work?
Nodes are a user-friendly approach to building conversations, because they allow both conversation designers and VA developers to conceptualize and visualize the entire flow of a conversation. Placed on a canvas, nodes help create a kind of conversation map that facilitates understanding for both the humans and the VA being developed.
Transitions add logic to the map created by the nodes. This way, the conversation can flow according to particular conditions, workflows or processes.
Both nodes and transitions work together, in context, to facilitate the creation of Virtual Assistants that can automate a wide variety of tasks, without taking away from the user or customer experience.
You can add the following nodes to dialog tasks, based on your requirements:
User Intent Node
The user intent to be identified by the Platform is based on the user utterance. Every dialog has one root intent with any number of sub-intents. For more information, refer to Working with the User Intent Node.
Dialog Task Node
The Dialog Task Node allows you to start a new dialog task within an existing one, if the user intent changes. Read more about the Dialog Task Node.
Entity nodes detect the entities within the user utterance, or ask the user for input when an entity is required. The user entity in the utterance. The platform supports 15+ entity types. Developers can define the prompt message to be shown, and this message can be channel-specific. For example, an amount, a flight account number, time, or zip code. Entity nodes prompt the user for an input. For more information, refer to Working with the Entity Node.
Form nodes allow you to integrate Digital Forms with Dialog Task by presenting a UI Form in the dialog task and making the form available for end-users on specific communication channels. This node is useful when you want to make entity collection more efficient by displaying a form rather than asking users for one entity at a time. Learn more about the Form node.
Displays a query message from the bot to the user, and then waits for a user response. Conditions are defined to continue processing the dialog task flow based on the user input. For more information, refer to Working with the Confirmation Node.
Message (or Bot Response) Node
Displays a message from the bot to a user, such as, Is there anything else I can help you with? For more information, refer to Working with the Message Node.
While any dialog starts at the Intent Node, it is advisable to end it with a Message Node to give a sense of closure to the user as well as to the bot internally.
Bot Action Node
The Bot Action Node can be used for any action that the Virtual Assistant is expected to perform without interaction with the user. Adding a Bot Action Node allows you to use five additional nodes, through which you can add a service, script, webhook, logic or process, using dedicated nodes.
Allows you to call an API. You can also use cURL to build the API request. For more information, refer to Working with the Service Node.
The Logic Node allows developers to use context variables and define complex transition conditions within Bot Action nodes and Dialog Tasks overall. For more details, please refer to Working with the Logic Node.
This node is used for server-side validation, executing business logic, or making backend server API calls. To use this node, you must have installed the SDK Tool Kit. For more information, refer to Working with the WebHook Node.
Process nodes allow you to connect a Dialog Task to a Process App that you have built within the Kore.ai XO Platform. Learn more about Process Nodes.
Agent Transfer Node
This node type is used to transfer communication from the bot to a live agent. This is usually the last node for a dialog task. For more information, refer to Working with the Agent Transfer Node.
A Dialog Task builds a task flow with different nodes connected by transitions. Dialog Task transitions depend on conditions that use business-defined evaluation criteria to take the next steps within the Dialog Task flow.
In the Kore.ai XO Platform, you define If-then-else conditions using a set of predefined operators. For each component, you must specify a fallback condition. How you represent a transition depends on the component type you are defining, and component type-specific transitions are described in the documentation topic for each component type.
You can use the following operators to write conditions:
- Does not exist
- Equals to
- Greater than equals to
- Less than equals to
- Not equals to
- Greater than
- Less than
Learn more about setting up node connections.
Environment Variables are global, reusable components which can be used in multiple places to define the bot configuration. Every Environment Variable is a key-value pair. The Variable Keys can be used at various places in the bot configuration. The platform will resolve/replace the Variable Keys with Variables Values during the conversation. Some of the typical use cases where a developer can use environment variables are:
- To manage the bot across environments having different endpoint URLs. The bot can be exported and imported based on the environment, and the infrastructure team will need to modify only the variable configuration file.
- To manage Conditions in the Dialog Node Transitions. Only the left-hand side of the transition condition uses Environment Variables. The right-hand side of the transition conditions does not use Environment Variables
- To define and manage the bot response that is repeated within the bot configuration as variables.
- Authorization Token
- Channel-related tokens and URLs
Post the release of v8.1 of the platform, for on-prem installations, multiple value sets can be stored for these environment variables using Collections, refer here for more.