The Help Center provides you with how-to's and tutorials to build your own conversational AI experiences. In addition, Cognigy provides a comprehensive developer documentation for a deep dive into the platform and details on each and every aspect of Cognigy.AI.
Cognigy.AI developer documentation
This list provides definitions of variouis Cognigy.AI terms. Follow the links to get more details in the developer documentation.
|Agents||Agents contain collections of conversational AI resources, including Flows, Lexicons, Endpoints and more. It is possible to work on a given Agent with multiple team members and Agents can deployed on a wide range of channels.|
|Analytics||Analytics provides an overview of message count, session count, unique users, misunderstood messages and execution time for the last 24 hours, week and month. Additionally, it also provides information about channel and Flow usage, as well as the top Intents in the last 30 days.|
|Connections||Connections are encrypted credentials that can be used with Extensions. The Connections view can be used to update and delete existing Connections. New Connections are created directly within the Extension Nodes that use them.|
|Contact Profiles||Contact Profiles store information about the end-users of your AI and can be accessed by Flows and Endpoints. Contact Profiles can be used to store information persistently and personalize discussions with users.|
|Deploy||Virtual Agents need to be deployed to channels in order for them to be available to the general audience. This process uses so-called Endpoints.|
|Endpoints||Endpoints help connect Channels (like Messenger or Alexa), to specific Cognigy Flows. They make sure that the channel-specific input and output (Images, Galleries, Voice) gets converted to the standardized Input object. This way, a Cognigy Flow needs to be developed only once, after which it can be deployed to many different channels by configuring Endpoints.|
|Flows||Flows provide a visual representation of a potential dialog in a user-friendly interface.
Flows are composed of individual Nodes with different functionalities, ranging from generating output to collecting back-end information, and are executed as soon as an Input object comes in through an Endpoint.
The Intent Trainer enables you to analyze the collected user inputs (records) and add them to Intents to improve your Agent's user input comprehension.
|Journeys||Journeys are on-screen tutorials that are part of the Help Center and provide step-by-step instructions, directly within the Cognigy.AI user interface.|
|Lexicons||Lexicons are collections of domain-specific Keyphrases (also known as Entities) that can be attached to a Flow. A Lexicon can be seen as a dictionary, that allows the Virtual Agent to "understand" specific words, like car brands, product groups or zipcodes. As soon as a Keyphrase is detected, it is published to the Input object for further use. This process is called Slot Mapping.|
|NLU Connectors||Cognigy.AI features built-in support for a number of third-party NLU engines.|
|Localization||Cognigy.AI features a powerful Localization concept, that allows for customizability combined with content re-use. The Localization view allows for the configuration of any number of localizations. The system works with fallback layers, meaning that a Locale that has not any content configured, will fallback to another Locale that has.|
The Logs allow you to track each user input and output in your Agent.
Here you can manage the Roles that people in your organization have in the project.
Cognigy.AI exposes an OData v4 analytics endpoint to retrieve analytics records and conversations.
|Playbooks||Playbooks help you test your Flows and make sure that they works as intended. They are automated conversations, which include Assertions that can check various elements of your Flow.|
|Settings||Cognigy.AI provides creators with customization settings that allow the agent processes to be adjusted to achieve optimal performance.|
|Snapshots||A Snapshot is an exported Agent. It can be used to transfer an Agent from one environment to another as they hold all relevant information. This way, no more individual resources have to be exported.|
|Tasks||Certain activities, like training Intents, importing Snapshots or the creation of a new Agent, generate so-called Tasks. A Task is an asynchronous process that can run in the background and is tied to the notification system.|
|Tokens||Tokens allow your advanced Cognigyscript users to provide simple and easy-to-use shortcuts to advanced functionality for your business users.|
|Tweak||One of the characteristics of Virtual Agents is that they can improve over time. In addition to a Self-learning mechanism, Cognigy.AI also provides the ability to tweak Agents based on existing dialogs.|