Summary metrics of one fitted model — model

The model-level metrics of evaluate_topic_models() for a single model: means over topics of coherence and exclusivity, topic diversity, mean and maximum similarity between topics, and the ldatuning criteria. Held-out likelihood and residuals need a held-out sample and are computed by evaluate_topic_models().

Usage

model_metrics(
  model,
  documents,
  metrics = c("coherence", "exclusivity", "diversity", "similarity", "ldatuning"),
  n_terms = 10L,
  frexw = 0.7,
  n_diversity = 25L,
  ...
)

Arguments

model: A fitted stm model (or a list with beta$logbeta[[1]] and vocab).
documents: The documents the model was fitted on (stm format).
metrics: Character vector among "coherence", "exclusivity", "diversity", "similarity", "ldatuning".
n_terms: Integer scalar, number of top words per topic (M in stm).
frexw: FREX weight of exclusivity.
n_diversity: Number of top words used for diversity (25 in Dieng et al. 2020).
...: Not used: an argument renamed in the development version of mintR gives an error naming its replacement.

Value

A one-row data.table.