Citation Patterns

Classify papers by the shape of their citation trajectory over time.

Citation Patterns

The Citation Patterns widget.

Overview

Classifies each paper’s citation trajectory into archetypal patterns (early rise, late bloomer, flash-in-the-pan, sustained, sleeping beauty, …) from its yearly citation history, and visualises the trajectories and their distribution. Works best with OpenAlex yearly-citation data.

Inputs

  • Data (Table) — data with yearly citation counts.

Outputs

  • Classified (Table) — each paper with its assigned pattern.

  • Selected Documents (Table) — papers selected in the plot.

Controls

  • Use OpenAlex API (recommended) — fetch yearly citations if not present.

  • Metric / Pattern / Plot Type — what to measure, which pattern to highlight, and how to display it (trajectories, distribution, …).

  • Metric bins — binning for the distribution view.

  • By Year from / to (0 = auto) — restrict the time window.

  • Normalize trajectories — scale each curve to 0–1 to compare shapes.

  • Distribution: single colour / sort by size, Metrics: show median line — display options.

  • Max papers, Min age (years) — sampling limits.

  • Auto apply — recompute on change.

Actions: Run Analysis, Export, Clear Selection.