Compute AOI sequence distance — compute_gazepoint_sequence

Computes a lightweight edit distance between two AOI sequences using a vector-based Levenshtein distance. This provides a simple scanpath dissimilarity measure without requiring heavy sequence-analysis dependencies.

Usage

compute_gazepoint_sequence_distance(
  sequence_a,
  sequence_b,
  ignore_missing = TRUE,
  missing_label = "missing",
  collapse_repeats = FALSE,
  substitution_cost = 1,
  insertion_cost = 1,
  deletion_cost = 1
)

Arguments

sequence_a: Character, factor, or atomic vector representing the first AOI sequence.
sequence_b: Character, factor, or atomic vector representing the second AOI sequence.
ignore_missing: Logical. If TRUE, missing and empty labels are removed.
missing_label: Character scalar used when ignore_missing = FALSE.
collapse_repeats: Logical. If TRUE, consecutive identical labels are collapsed before distance is computed.
substitution_cost: Numeric scalar substitution cost.
insertion_cost: Numeric scalar insertion cost.
deletion_cost: Numeric scalar deletion cost.

Value

A one-row data frame with edit distance, normalized distance, and sequence lengths.

Examples

compute_gazepoint_sequence_distance(
  sequence_a = c("Claim", "Evidence", "CTA"),
  sequence_b = c("Claim", "CTA", "Evidence")
)
#>   edit_distance normalized_distance sequence_a_length sequence_b_length
#> 1             2           0.6666667                 3                 3
#>   distance_status
#> 1              ok