JevMade

Sign in
← Back to guides

JevMade field notes / Worked recipe

Add searchable labels to product descriptions

This recipe turns product descriptions into categories and tags that a database can search, while leaving prices and exact comparisons to ordinary software.

Original by Jeroen Erne / NexibeoClassification
Leave a noteWhat would you try with this? : Add searchable labels to product descriptions

Only you can see your notes.

Listen to this guide

JevMade’s plain-English explanation

0:00 /

Our summary

A shop's descriptions may suggest that an item is a gift or suits a certain audience without storing those details separately. This recipe asks Jev to identify categories, audiences, and tags, then saves its answers alongside the unchanged product records.

The program checks which products still need labels before sending requests. It stores categories, probabilities, the model used, and the indexing date in separate tables. The database can then combine those labels with exact conditions, such as a maximum price, without asking AI to do arithmetic.

Some labels in the author's sample were debatable or wrong. A saved probability helps someone find uncertain records but does not prove the other labels are correct. Clear tag definitions matter, and changed questions or model versions may require checking and labelling products again.

Key takeaways

  1. Preserve the original records when adding inferred labels.
  2. Store the model, date, and confidence with each judgment.
  3. Use database rules, not AI, for prices and arithmetic.

Your note

Only you can see this. Clear the text and save to remove it.

Keep what you save

Sign in to keep bookmarks and private notes on every device. We email you a link; there is no password, and the same link creates your account.