Skip to content

Sorting with Help from Machine Learning #1422

Description

@OlegPhenomenon

To boost user engagement, we could implement sorting based on machine learning models (not to be confused with language models!). The idea is to build a pipeline that tags each new domain with relevant labels—basically adding some meaningful context to each one. This tagging part can actually be handled by a language model.

Then, we track each user’s activity—what kind of domains they’re bidding on. Using pre-trained machine learning models (no need to build one from scratch!), we can categorize domains and show the most relevant ones at the top of the list for each user.

For first-time users, we could start with a short quiz. We’d present it as a way to “personalize your auction feed.” For example, if someone wants to do dropshipping, they might prefer a used domain (which is great for SEO) and one related to online stores. They can tell us that in the quiz—and boom, we show them domains that match that theme more often when they browse the auction.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions