Skip to main content

Tag Acceptance Models

When integrating Geneea into a Content Management System (CMS), one of the most important decisions is how journalists will interact with the suggested tags.

We categorize these workflows into three models: Opt-in, Opt-out, and Mixed.

1. Opt-in Model​

In this mode, tags are suggested, but none are applied to the article until a journalist manually selects them.

  • How it works: The CMS displays a list of suggested tags. The journalist clicks the ones they want to keep.
  • Filtering: Geneea system is configured to return tags above certain threshold; The tags might b
  • Best for: High-stakes editorial environments where tagging accuracy is more important than speed, or where tag taxonomies are very strict.
  • Pros: 100% human-verified; no "weird" tags ever reach the public site.
  • Cons: Higher click-debt for journalists; tags may be forgotten if the editor is in a hurry.

2. Opt-out Model​

In this mode, all the returned tags are automatically applied to the article by default; the journalist might reject them.

  • How it works: When the article is analyzed, the tags appear in the "Active Tags" area immediately. The journalist only interacts with them if they see something they want to remove.
  • Best for: High-volume newsrooms, breaking news, or SEO-heavy workflows where maximum coverage is the priority.
  • Pros: Extremely fast; ensures every article is tagged even if the editor does nothing.
  • Cons: Higher risk of "false positives" reaching the live site.

This is our recommended approach. It uses Relevance Scores to split tags into two categories: Confident and Suggested.

  • How it works:
    • Auto-Accept (Relevance > 80): These tags are applied automatically (Opt-out style).
    • Suggestions (Relevance 40 – 80): These are shown in a "Recommended" bin for the journalist to click and add (Opt-in style).
    • Hidden (Relevance < 40): These are ignored or only used for background search indexing.
  • Best for: Most professional newsrooms.
  • Pros: Automates the "obvious" work while leaving the nuanced decisions to the humans. It drastically reduces the number of clicks required while maintaining a high quality bar.

Comparison Summary​

ModelSetupSpeedEditorial Control
Opt-inAll tags are "Suggestions"LowMaximum
Opt-outAll tags are "Applied"HighMinimum
Mixed*High relevance = Applied; Medium = SuggestionHighBalanced

Technical Implementation​

Presentation in the UI​

To implement these models effectively, your CMS interface should visually distinguish between the states.

  1. The "Active" Bin: Tags already attached to the article. (In Opt-out and Mixed modes, this is pre-populated).
  2. The "Suggestions" Bin: A cloud or list of tags that can be "promoted" to Active with a single click.

Closing the Loop with Feedback​

Regardless of the model you choose, we highly recommend sending the final state of the tags back to our API.

If a journalist removes an auto-applied tag (Opt-out) or ignores a high-relevance suggestion (Opt-in), that is a valuable signal. Sending this Feedback helps us understand your specific editorial preferences and can be used to tune your results over time.