Field Networks

Jaccard heatmap, disparity-filter backbone and bridging nodes for field co-occurrence.

Field Networks

The Field Networks widget.

Overview

Advanced co-occurrence analysis for OpenAlex fields/subfields (or any multi-valued column): a normalised (Jaccard / association / …) field × field heatmap, a disparity-filter backbone that keeps only statistically strong links, Louvain communities and bridging centralities (nodes linking otherwise separate areas). Emits Node/Edge tables that plug into Plot Bibliometric Network.

Inputs

  • Data (Table) — data with a multi-valued field (oa_fields/subfields, keywords, countries…).

Outputs

  • Node Data (Table) — community, degree, betweenness, clustering, bridging.

  • Edge Data (Table) — Source/Target/Weight (feed the plot widget).

  • Bridging Nodes (Table) — top bridging nodes.

Controls

  • Column — the multi-valued field analysed.

  • Min entity count — drop rare entities.

  • Normalisation — jaccard / association / inclusion / salton / none.

  • Min edge weight — prune weak links.

  • Disparity-filter backbone + Backbone alpha — keep only significant edges (smaller α = stricter).

  • Heatmap top N / Bridging top N — sizes of the two views.

  • Run automatically — recompute on change.

Actions: Run.