Skip to content

Add tier weighting to influence prediction #9

@benevolentbandwidth-nurtekin

Description

As we expand across research domains, not all venues and publication contexts carry the same signal strength. We want to incorporate tier-based weighting into the impact prediction model (e.g., conferences, journals, venues).

Goal:

Explore whether venue tiers (or similar structured signals) can improve the robustness and generalization of impact prediction.

Possible areas to consider:

  1. Conference vs journal weighting
  2. Venue reputation tiers
  3. Configurable weighting multipliers
  4. Ensuring fairness across different research fields

Key Considerations

  1. Avoid hardcoding rigid or biased rankings
  2. Keep weighting configurable where possible
  3. Ensure changes improve prediction stability, not just ranking volatility

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions