Cluster decision paths

Groups casualties by how similar their board pathways are: an optimal matching dissimilarity between state sequences, then partitioning around medoids. What comes out is a small number of characteristic ways through the board, which is usually more useful than a per-casualty path listing.

Usage

sdt_seq_cluster(run, k = 3L, method = "OM", indel = 1, sm = "CONSTANT")

Arguments

run

A game_run from sdt_simulate().

k

Number of clusters.

method

Dissimilarity passed to TraMineR::seqdist(). "OM" is optimal matching.

indel

Insertion/deletion cost for optimal matching.

sm

Substitution-cost matrix or method. "CONSTANT" applies a flat cost, the right default when no substitution is more plausible than another.

Value

A tibble::tibble() with one row per sequence: replication, id, cluster, and medoid, which flags the sequence representing its cluster.

See also

sdt_seqdef(), sdt_plot_sequences().

Examples

co <- sdt_synthetic_cohort(n = 30, seed = 1)
run <- sdt_simulate(sdt_scenario(co), replications = 2, seed = 1,
                    verbose = FALSE)
if (requireNamespace("TraMineR", quietly = TRUE) &&
    requireNamespace("WeightedCluster", quietly = TRUE)) {
  sdt_seq_cluster(run, k = 3)
}
#> Registered S3 method overwritten by 'vegan':
#>   method          from   
#>   print.nullmodel parsnip
#> # A tibble: 60 × 4
#>    replication id    cluster medoid
#>          <int> <chr>   <int> <lgl> 
#>  1           1 P001        2 FALSE 
#>  2           1 P002        2 FALSE 
#>  3           1 P003        3 FALSE 
#>  4           1 P004        1 FALSE 
#>  5           1 P005        2 FALSE 
#>  6           1 P006        1 FALSE 
#>  7           1 P007        1 FALSE 
#>  8           1 P008        3 FALSE 
#>  9           1 P009        3 FALSE 
#> 10           1 P010        2 FALSE 
#> # ℹ 50 more rows