Skip to contents

build_cna() estimates an undirected co-occurrence network (CNA): for every pair of positions in a sequence where both are observed, the two states count as co-occurring once, in either order. It is build_network() with method = "co_occurrence", and takes sequences in a wide table, an event log, one-hot codes (declared with codes or format = "onehot", and counted in windows as wtna() does) or a square matrix of counts.

An edge counts how often two states occur together within sequences (over pairs of positions, so a sequence can add to one edge several times), not which came first; for order, use build_tna(). Check how far the counts move under resampling with bootstrap_network().

Usage

build_cna(
  data,
  start = FALSE,
  end = FALSE,
  actor = NULL,
  action = NULL,
  time = NULL,
  session = NULL,
  ...
)

Arguments

data

The sequences: a data frame with one row per sequence and one column per position, an event log (one row per event, with action set), one-hot code columns, or a square matrix of co-occurrence counts (non-negative whole numbers). Column roles are declared as in build_network().

start

A start state put before every sequence, so the network shows where sequences begin. FALSE (default) adds none, TRUE adds one labelled "Start", a single string adds one with that label.

end

An end state put after each sequence's last observed state, so the network shows where sequences stop; it has no outgoing edges. FALSE (default), TRUE ("End") or a single string, as start.

actor

The column naming who each row belongs to (a person, a team). In an event log, each actor's events form a sequence, split further by session or time_threshold. Default NULL.

action

The column holding each event's state; setting it declares the data an event log. Column names give no role: action, actor, time and session are read only when set. A column called Action with nothing declared raises an error of class nestimate_bad_input, and an undeclared time or session column beside action is named in a message of class nestimate_column_role. Default NULL.

time

The timestamp column of an event log. Events are ordered by it, and a gap longer than time_threshold starts a new sequence. Default NULL.

session

The column identifying sessions in an event log; each session is its own sequence, also without actor. Default NULL.

...

Further arguments for build_network(), such as group, params, scaling, threshold or state_cols. method cannot be given: it is always "co_occurrence" (error of class nestimate_bad_input).

Value

A netobject (a netobject_group with group), exactly as build_network() returns it: print it for an overview, read the edges with extract_edges() and the weights with extract_transition_matrix(), and draw it with cograph::splot().

See also

build_network, cooccurrence for delimited-field, bipartite, and other non-sequence co-occurrence formats.

Examples

net <- build_cna(trajectories)
net
#> Co-occurrence Network [undirected]
#>   Weights: [663.000, 3179.000]  |  mean: 2134.000
#> 
#>   Weight matrix:
#>              Active Average Disengaged
#>   Active       3156    2783        663
#>   Average      2783    3179       2075
#>   Disengaged    663    2075        948