Select to view content in your preferred language

AI-Powered Metadata Quality, Dependency Analysis, and Content Discovery Assistant for ArcGIS Online

75
1
Tuesday
Status: Open
ShreedatriDey_agolidea
Emerging Contributor

As an ArcGIS Online administrator, I need a simple way to assess metadata quality, understand item relationships, and discover relevant content across the organization. Today, identifying which maps use specific layers, understanding group layer hierarchies, or finding content with poor metadata requires significant manual effort.

An AI-powered assistant integrated with Semantic Search and a chat interface could analyze metadata, dependencies, ownership, tags, and usage patterns to answer questions such as:

  • Which maps use this layer?
  • What content has incomplete metadata?
  • Which items are duplicates or rarely used?
  • What would be impacted if this layer is removed?

This capability would improve content governance, metadata quality, discoverability, and administrative efficiency, helping organizations manage ArcGIS Online content at scale.

1 Comment
RichardDaniels

Using AI to address an interface design failure would be an interesting solution. It could work if the API was fully documented and functional so the AI generated code could obtain the needed data from the underlying ArcGIS Online system and imbedding the needed credentials for each organization is simple and well documented.

However, I much rather have ESRI spend some resources to build a functional Admin console for ArcGIS Enterprise and ArcGIS Online that focuses on Administrative tasks and subscription (a.k.a., license) management. We spend thousands on the AGOL capabilities but still have no or very limited administrative tools. We need to be able to see usage levels and frequencies to make intelligent decisions on subscription distribution and levels (not everyone can be Professional Plus 😉). This frequency and usage analysis is available for services but the way you access the graphs is very limited, you can only see ONE at a time which makes comparison nearly impossible.

Sincerely,

Rich