Premise x neighbourhood co-occurrence network

Builds a co-occurrence network in lieu of the co-offender network from D'Orsogna & Perc (2015) Fig. 9 / Diviak et al. (2019). Public TPS data has no co-offender records, so we approximate by projecting (top-N premise types) x (HOOD_158 neighbourhoods) onto a premise-by-premise edge-weighted graph. Edge weight is the count of neighbourhoods in which both premise types appear.

Usage

morie_tps_criminal_network_graph(
  category = "Assault",
  sample_rows = 30000L,
  top_n_premises = 20L,
  save_fig = TRUE,
  fig_dir = NULL
)

Arguments

category

TPS category name.

sample_rows

Maximum rows to load.

top_n_premises

Number of premise nodes to keep.

save_fig

Whether to emit a circular layout PNG.

fig_dir

Directory to write the PNG into; NULL (the default) skips writing and says so in the result.

Value

A morie_rich_result with node count, edge count, strongest edge weight, and the adjacency payload.

References

Diviak T, Dijkstra JK, Snijders TAB (2019). Structure, multiplexity, and centrality in a corruption network. Trends in Organized Crime 22: 274-297.

Examples

# \donttest{
  rr <- morie_tps_criminal_network_graph("Assault",
                                           top_n_premises = 10L,
                                           save_fig = FALSE)
#> morie_tps: no local TPS cache for 'Assault'; using the bundled sample from rmoriedata. Fetch the full export with morie_tps_fetch_category().
  print(rr$summary_lines)
#> $Method
#> [1] "Premise x neighbourhood co-occurrence network"
#> 
#> $Nodes
#> [1] 7
#> 
#> $Edges_pos
#> [1] 21
#> 
#> $max_edge_weight
#> [1] 89
#> 
#> $Figure
#> [1] "(skipped)"
#> 
# }