How To Tag Data In Relevance AI

How To Tag Responses in Relevance AI

After data is uploaded to the Relevance AI platform, tagging can be done through a variety of options. The most common tagging workflow is AI-Tagging.

This workflow allows you to use your own code frames/ tags or automatically extracts keywords/key phrases (or code frames) using AI models. Then you are able to review the results and apply the final list to your dataset. You are provided with a user-friendly GUI as well, which allows you to refine the tagging results.

Note1: it is highly recommended to include your domain knowledge by modifying the candidate tags as well as refining the results.

Note2: Tagging algorithms require complex computation that can take long on large datasets. You can start your workflow on the cloud and take care of other tasks while your dataset is being updated with tagging results.

Review: Edit, Update, Merge and Finalize Tags

Export Responses With Tags

If you wish to download your tags in a flat list CSV (i.e. one column per tag with True/False values if the entry is labelled with a tag), use the wide export format on Explorer, or the Export tags workflow.

Relevance AI - Export tags in flat/wide format

Relevance AI - Export tags in flat/wide format

If you wish to see the results first and explore for insights Explorer is the best tool. It is designed to extract insights from unstructured data, an advanced tool, fully configurable for different personas in a company.

The Explorer dashboard is easy to setup with variety of functionalities from the way the data is presented, to how you can search, filter and analyse the whole data or your desired subset.

Explorer allows you to present your data under categories based in the tagging results. Read more about the Explorer and specifically how to set up a category view. There is also detailed explanation on How to export data to CSV but tags in this export are not in a flat format.

TIPS - What is a good tag?

Is the tag precise?
This means that the tag properly conveys the intended topic and minimises opportunity for ambiguity.

There are several ways that a tag can be not precise. For example: “Pricing/dollars, incl. deals”

  • Could be split into 2 separate tags: “Pricing” and “Deals”

  • “Convenience of Site” → “Close Proximity”

  • "sydney" -> "Sydney" (Capital letters matter!) and similarly "Product Condition" -> "product condition" will help the AI

  • "Shopping Experience" could instead be broken down into different components that make up the shopping experience such as "checkout experience", "parking lot", "friendly staff".

Is the tag concise?
This means that this is not an incredibly wordy tag.

A tag should not have unnecessary words where possible, as this can dilute meaning.

  • For example: “Good Customer Service” → “Customer Service” (unless you are specifically looking for the good component of the word.
  • This is important because good customer service will not tag customer service but customer service will tag good customer service).

Related Articles (coming soon)

Tips to improve tag accuracy

Editing, Updating and Merging Tags, Categories or Themes

Tips To Improve Tag Accuracy and Granularity

What is the minimum number of records to get meaningful results?