Learn to set up your conversational AI application with NLX in minutes
Last updated
What's an Application?
A conversational AI application within the NLX platform is defined by the workflows attached to it and the channels where it's deployed to interface with human users.
Once an intent flow is attached to your application, the application becomes imbued with the ability to deliver whatever process or response the flow has designed (booking a spa appointment, ordering room service, etc.).
Since intents are modular and may be assigned to one or more applications in NLX, each application created may be unique based on the intents attached to it, language settings assigned, and channels deployed. Some workspaces may have 2-4 duplicate applications to serve different production, staging, or dev environments, while others may be specified based on particular customer service tasks or even region.
To access intents, click Applications in your workspace menu:
Requirements
An application must always employ the following for being built and deployed to the world:
An application's Settings tab allows you to update basic application information, set criteria for various application features, enable LLM-powered intent classification, and delete your application.
Open the application you wish to edit and select its Settings tab.
The Details section allows you to add or edit an optional description and resource tags.
The General section allows you to toggle on several important features for your application:
Child-directed: For applications subject to COPPA. Utterance information will not be stored if enabled
Autocorrection: For supported NLPs, spell correction will be applied towards written user responses
Repeat on incomprehension: If a user's response is unclear, the application will repeat its last question
Enable sentiment: Allows your application to detect moods or sarcasm from the user (if using Amazon Lex NLP with your application, be sure to enable sentiment in your console)
Enable NLX Boost: Enhances the performance of your NLP engine with LLM intent classification
Allow NLX Boost to override the NLP: Enabling this relies on NLX's built-in LLM to detect user intent (based on your training data) and route accordingly, regardless of NLP's detected match
By default, all features in this section are turned off.
NLP engines perform well when trained well. To set your desired level of confidence based on your training quality, we provide confidence thresholds that deem when the application should take the NLP response into account and when it should not:
NLP high confidence threshold: When an utterance falls above this value, your bot assumes it matched to an intent with certainty
Negative sentiment threshold: A frustration flow or response will be triggered when negative sentiment is detected (e.g., profanity, sarcasm, etc.)
NLX Boost confidence threshold: A custom threshold for when NLX Boost is enabled
Incomprehension count: Define the number of sequential utterances that the application asks for clarity on before an escalation or Unknown intent is triggered
Conversation timeout (min): Sets the timeout period for all channels on your application. If a timeout was also configured for a channel when managing channels, that channel's setting will take precedence over the application's setting
Default project ID: Default Project ID when using Google's Dialogflow NLP
Should you want to delete an application that's no longer in production, you may do so through the Danger zone.
Deleting an application does not delete the resources attached to your application, but the deletion of your application does wipe all custom settings configured for communication channels you may have set up for the application or any automated tests.
