HospiNet function

Class providing the HospiNet object with its methods

Class providing the HospiNet object with its methods

Class providing the HospiNet object with its methods class

Format

R6::R6Class object.

Returns

Object of R6::R6Class with methods for accessing facility networks.

Methods

• new(edgelist, window_threshold, nmoves_threshold, noloops): This method is used to create an object of this class with edgelist as the necessary information to create the network. The other arguments window_threshold, nmoves_threshold, and noloops are specific to the edgelist and need to be provided. For ease of use, it is preferable to use the function hospinet_from_subject_database().

• print(): This method prints basic information about the object.

• plot(type = "matrix"): This method plots the network matrix by default. The argument type can take the following values:

   - **matrix**: plot the network matrix,
   - **clustered_matrix**: identify and plot cluster(s) in the matrix using the infomap algorithm (from igraph),
   - **degree**: plot the histogram of the number of neighbors by facility,
   - **circular_network**: plot the network by clusters using a "spaghetti-like" layout. Only works when there are at least 2 clusters.
  

Examples

mydbsmall <- create_fake_subjectDB(n_subjects = 100, n_facilities = 10)

hn <- hospinet_from_subject_database(
  base = checkBase(mydbsmall),
  window_threshold = 10,
  count_option = "successive",
  condition = "dates"
)

hn

plot(hn)
plot(hn, type = "clustered_matrix")

Active bindings

• edgelist: (data.table) the list of edges (origin, target) and their associated number of movements (N) (read-only)

• edgelist_long: (data.table) edgelist with additional information (read-only)

• matrix: (matrix) the transfer matrix (active binding, read-only)

• igraph: (igraph) the igraph object corresponding to the network (active binding, read-only)

• n_facilities: the number of facilities in the network (read-only)

• n_movements: the total number of subject movements in the network (read-only)

• window_threshold: the window threshold used to compute the network (read-only)

• nmoves_threshold: the nmoves threshold used to compute the network (read-only)

• noloops: TRUE if loops have been removed (read-only)

• hist_degrees: histogram data of the number of connections per facility

• LOSPerHosp: the mean length of stay for each facility (read-only)

• admissionsPerHosp: the number of admissions to each facility (read-only)

• subjectsPerHosp: the number of unique subjects admitted to each facility (read-only)

• degrees: number of connections for each facilities (total, in, and out)(read-only)

• closenesss: the closeness centrality of each facility (read-only)

• betweennesss: the betweenness centrality of each facility (read-only)

• cluster_infomap: the assigned community for each facility, based on the infomap algorithm (read-only)

• cluster_fast_greedy: the assigned community for each facility, based on the greedy modularity optimization algorithm (read-only)

• hubs_global: Kleinberg's hub centrality scores, based on the entire network (read-only)

• hubs_infomap: same as hubs_global, but computed per community based on the infomap algorithm (read-only)

• hubs_fast_greedy: same as hubs_global, but computed per community based on the infomap algorithm (read-only)

• metricsTable: (data.table) all of the above metrics for each facility (read-only)

Methods

Public methods

• list("HospiNet$new()")

• list("HospiNet$print()")

• list("HospiNet$plot()")

• list("HospiNet$clone()")

Method new()

Create a new HospiNet object.

Usage

HospiNet$new(
  edgelist,
  edgelist_long,
  window_threshold,
  nmoves_threshold,
  noloops,
  prob_params,
  fsummary = NULL,
  create_MetricsTable = FALSE
)

Arguments

• edgelist: Short format edgelist

• edgelist_long: Long format edgelist

• window_threshold: The window threshold used to compute the network

• nmoves_threshold: The nmoves threshold used to compute the network

• noloops: TRUE if loops have been removed

• prob_params: Currently unused

• fsummary: A pre-built data.table with the LOSPerHosp, subjectsPerHosp and admissionsPerHosp that don't need to be recomputed.

• create_MetricsTable: all of the metrics for each facility

Returns

A new HospiNet object

Method print()

Prints a basic description of the number of facilities and movements of a HospiNet object.

Usage

HospiNet$print()

Returns

NULL

Method plot()

Plots various representations of the HospiNet network

Usage

HospiNet$plot(type = "matrix", ...)

Arguments

• type: One of "matrix", "degree", "clustered_matrix", "circular network" Choose what you would like to plot - the connectivity matrix, degree distribution, the clusters, or the network in a circle.

• ...: Additional arguments to be provided. Only supported for type == 'circular_network'.

Returns

a ggplot2 object

Method clone()

The objects of this class are cloneable with this method.

Usage

HospiNet$clone(deep = FALSE)

Arguments

• deep: Whether to make a deep clone.