sfa_nameability: Nameability of an Orientation

In semanticfa: Semantic Factor Analysis of Language Model Embeddings

Nameability of an Orientation

Description

Runs the full sfa_name() selection rule (family-gated candidate walk, construct-noun preference, leave-one-out sets, geometric collision keeper) on an arbitrary pattern matrix of the lexical map's items, and returns the retrieved labels together with the nameability criterion: per factor, the cosine between the factor's naming target and its retrieved term's embedding, and their mean.

Usage

sfa_nameability(
  lexmap,
  loadings,
  n_candidates = 5L,
  collision = TRUE,
  loo_sets = TRUE,
  baseline = NULL,
  baseline_sd = NULL
)

Arguments

lexmap	An sfa_lexmap() object.
loadings	A pattern matrix (items x factors) in any orientation of the fitted solution.
n_candidates	Visible label window (default 5, as in sfa_name()).
collision	Resolve duplicate labels via the geometric keeper.
loo_sets	Compute leave-one-out candidate sets.
baseline	Optional numeric vector, one value per pool word, giving that word's expected target cosine under a null. When supplied, word scores are centered on it ("surprise scoring"), which suppresses words that sit close to every target rather than to this factor in particular. Default NULL leaves scores uncentered.
baseline_sd	Optional numeric vector, one value per pool word, giving the standard deviation of that word's null target cosine. When supplied together with baseline, scores are standardized rather than merely centered. Default NULL.

Value

A list with labels (data frame: factor, label, rule, n_items, collision_moved, row), candidates (list column), criterion (mean), and per_factor.

semanticfa documentation built on Sept. 2, 2026, 1:07 a.m.