Rank and Select the Top Similar Alternative Sites

Orders candidate sites by similarity score (highest first) and returns the top n_select, annotated with rank and target site information.

Usage

ss_alt_rank(
  similarity_scores,
  candidate_sites,
  n_select = 5,
  target_site_id = NULL
)

Arguments

similarity_scores

Numeric vector of similarity scores, as returned by ss_alt_similarity(), one per row of candidate_sites.

candidate_sites

Data frame of candidate sites, as returned by ss_alt_candidates() or ss_alt_filter_buffer().

n_select

Integer, number of top alternatives to select. If fewer valid (non-NA, finite) scores are available, all of them are returned with a warning.

target_site_id

Optional character, ID of the target site these alternatives are for; stored in a target_site_id column. If NULL (default), that column is filled with NA.

Value

A data frame of the top n_select candidate sites, with added target_site_id, similarity_score, similarity_rank, and selection_method columns. Empty data frame if no valid scores.

See also

ss_alt_similarity(), ss_alt_sites()

Examples

candidates <- data.frame(site_id = paste0("c", 1:5), x = 1:5, y = 1:5)
scores <- c(0.9, 0.4, 0.95, 0.2, 0.7)
ss_alt_rank(scores, candidates, n_select = 2, target_site_id = "site_001")
#>   site_id x y target_site_id similarity_score similarity_rank selection_method
#> 3      c3 3 3       site_001             0.95               1       similarity
#> 1      c1 1 1       site_001             0.90               2       similarity