Adding Metadata to Knowledge Articles

This topic describes how you can use content tag metadata to filter Knowledge Articles after they have been ingested to the Aisera Gen AI Platform.

Before Ingestion into the Aisera Gen AI Platform

When you are planning to ingest Ticket and Knowledge Base Article data from your Enterprise software to the Aisera Gen AI platform, you may want to add metadata tags to your articles before connecting your Enterprise data store to the Aisera platform.

The instructions for adding tags to your articles should be in the Product Documentation for the system you are using.

For example:

If you are using Zendesk, information about how to add content tags is here: https://support.zendesk.com/hc/en-us/articles/5003635888410-Adding-content-tags-to-group-related-articles

For Salesforce, the information is here:

https://trailhead.salesforce.com/trailblazer-community/feed/0D54S00000A8J5ZSAV

For ServiceNow, the information is here:

https://www.servicenow.com/community/servicenow-ai-platform-forum/is-is-possible-to-apply-tags-to-knowledge-articles/m-p/1143884

For Confluence, see: Ingesting Dynamically Generated Content from Confluence (adds new field)

For NexThink, the information is here: https://docs.nexthink.com/platform/user-guide/administration/content-management/custom-fields-management

You may need your company’s administrator for your Enterprise system to add these content tags.

How Content Tag Metadata Can Be Used in the Aisera Gen AI Platform

You can limit which users see which content as fulfillment in an application/bot by adding content tags.

Contant Access Control

Content Tags are associated with user permissions, since you are filtering by the user and the topic tags.

The Aisera Gen AI platform offers two types of content access control:

Content-level Permissions (ACLs per Knowledge Article)

Content Access Controls via APIs (if supported)

Channel Metadata Filters

You can limit which content is returned to a specific channel by setting up filters when you add a channel to your tenant configuration. Specify metadata tags as key-value pairs during channel configuration for use in RAG searches. When a RAG search is invoked, the metadata tags are appended to the end of the user’s query, allowing different channels to return tailored results while still utilizing the same bot.