Title: Represent Network Objects on a Map
Version: 0.1.4
Description: Represent 'network' or 'igraph' objects whose vertices can be represented by features in an 'sf' object as a network graph surmising a 'sf' plot. Fits into 'ggplot2' grammar.
License: GPL (≥ 3)
Encoding: UTF-8
RoxygenNote: 7.3.1
URL: https://github.com/artod83/netmap
BugReports: https://github.com/artod83/netmap/issues
Imports: ggnetwork, igraph, network, rlang, sf, sna
Suggests: rmarkdown, knitr, testthat (≥ 3.0.0)
Config/testthat/edition: 3
VignetteBuilder: knitr
Depends: R (≥ 2.10)
LazyData: true
NeedsCompilation: no
Packaged: 2024-02-26 13:25:09 UTC; 146307
Author: Matteo Dimai ORCID iD [aut, cre]
Maintainer: Matteo Dimai <matteo.dimai@phd.units.it>
Repository: CRAN
Date/Publication: 2024-02-26 13:50:10 UTC

netmap: Represent Network Objects on a Map

Description

Represent 'network' or 'igraph' objects whose vertices can be represented by features in an 'sf' object as a network graph surmising a 'sf' plot. Fits into 'ggplot2' grammar.

Author(s)

Maintainer: Matteo Dimai matteo.dimai@phd.units.it (ORCID)

See Also

Useful links:

Internal checks before ggnetmap and ggcentrality

Description

Checks whether the proper packages are installed, whether the parameters are of the proper classes, whether the network-map link is possible, then performs the link.

Usage

check_network_sf(n, m, lkp = NULL, m_name = NULL, n_name = "vertex.names")

Arguments

n

A network or igraph object.

m

A sf object.

lkp

An optional lookup table.

m_name

Optional character, name of field in m and of column in lkp.

n_name

Optional character, name of vertex attribute in n and of column in lkp.

Value

A list with a network or igraph object with only the vertices present in the sf object as the first element and a list with two vectors, one of features in m present both in the lookup table and in n, the other of nodes in n present both in the lookup table and in m

Map of municipality borders in the Friuli Venezia Giulia region, Italy

Description

An sf object containing the ISTAT municipality codes, geometry and the municipality names in the Friuli Venezia Giulia region in northeastern Italy, based on official ISTAT shapefiles.

Usage

fvgmap

Format

An sf object with 215 features and 6 fields:

Cod_reg

region code, always =6 (Friuli Venezia Giulia)

Cod_pro

province code (93=Pordenone, 30=Udine, 31=Gorizia, 32=Trieste)

Pro_com

municipality code, consists of province code + progressive code of the municipality within the province

Shape_leng

length of municipality perimeter

Shape_area

municipality area

geometry

a MULTIPOLYGON

Source

https://www.istat.it/it/archivio/104317

Calculate centrality indices for vertices linked to a sf object

Description

Given a sf object with features that can be linked to a network or igraph object, obtain centrality indices for linked features.

Usage

ggcentrality(
  n,
  m,
  lkp = NULL,
  m_name = NULL,
  n_name = "vertex.names",
  par.deg = NULL,
  par.bet = NULL,
  par.clo = NULL
)

Arguments

n

A network or igraph object.

m

A sf object.

lkp

An optional lookup table.

m_name

Optional character, name of field in m and of column in lkp.

n_name

Optional character, name of vertex attribute in n and of column in lkp.

par.deg

List with additional optional parameters to functions degree or degree.

par.bet

List with additional optional parameters to functions betweenness or betweenness.

par.clo

List with additional optional parameters to functions closeness or closeness.

Value

An sf object, input m with added columns for centrality indices (degree, betweenness, closeness; existing columns with the same name will be overwritten) and with only the features linked to vertices in input n.

Examples

net=network::network(matrix(c(0, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0), nrow=4, byrow=TRUE))
network::set.vertex.attribute(net, "name", value=c("a", "b", "c", "d"))
wkb = structure(list("01010000204071000000000000801A064100000000AC5C1641",
"01010000204071000000000000801A084100000000AC5C1441",
"01010000204071000000000000801A044100000000AC5C1241",
"01010000204071000000000000801A024100000000AC5C1841"), class = "WKB")
map=sf::st_sf(id=c("a1", "b2", "c3", "d4"), sf::st_as_sfc(wkb, EWKB=TRUE))
lkptbl=data.frame(id=c("a1", "b2", "c3", "d4"), name=c("a", "b", "c", "d"))
netmap::ggcentrality(net, map, lkptbl, "id", "name")

Calculate connectedness to a specific vertex for vertices linked to a sf object

Description

Given a sf object with features that can be linked to a network or igraph object and given a node with id id in said graph that can be linked to the sf object, obtain an indicator variable denoting, for each node, a connection to id.

Usage

ggconn_area(n, m, id, lkp = NULL, m_name = NULL, n_name = "vertex.names")

Arguments

n

A network or igraph object.

m

A sf object.

id

The identifier (as vertex attribute n_name of object n) of the feature that needs to be checked for connections.

lkp

An optional lookup table.

m_name

Optional character, name of field in m and of column in lkp.

n_name

Optional character, name of vertex attribute in n and of column in lkp.

Value

An sf object, input m with an added column conn_area with an indicator variable set to 1 if the feature is connected to the feature with vertex id id, 0 otherwise. In directed graphs, only outgoing links are considered a connection. Any existing column with the same name will be overwritten, the result will contain only the features linked to vertices in input. If the vertex id is not present in object n, conn_area will be set to 0 for all vertices.

Examples

net=network::network(matrix(c(0, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0), nrow=4, byrow=TRUE))
network::set.vertex.attribute(net, "name", value=c("a", "b", "c", "d"))
wkb = structure(list("01010000204071000000000000801A064100000000AC5C1641",
"01010000204071000000000000801A084100000000AC5C1441",
"01010000204071000000000000801A044100000000AC5C1241",
"01010000204071000000000000801A024100000000AC5C1841"), class = "WKB")
map=sf::st_sf(id=c("a1", "b2", "c3", "d4"), sf::st_as_sfc(wkb, EWKB=TRUE))
lkptbl=data.frame(id=c("a1", "b2", "c3", "d4"), name=c("a", "b", "c", "d"))
ggconn_area(net, map, "b", lkptbl, "id", "name")

Fortify a network over a map

Description

Link a network or igraph and a sf object in a data.frame for subsequent representation on a plot using ggplot2.

Usage

ggnetmap(
  n,
  m,
  lkp = NULL,
  m_name = NULL,
  n_name = "vertex.names",
  scale = FALSE,
  ...
)

Arguments

n

A network or igraph object.

m

A sf object.

lkp

An optional lookup table.

m_name

Optional character, name of field in m and of column in lkp.

n_name

Optional character, name of vertex attribute in n and of column in lkp.

scale

Whether coordinates should be scaled (defaults to FALSE since the network should be overlayed with the non-scaled sf object).

...

Additional parameters passed to fortify.

Details

Using a network or igraph and a sf object as inputs, with an optional lookup table (a data.frame) in case the IDs don't match, produces a data.frame that can be used with ggnetwork's geom_edges and geom_nodes functions to represent the network as overlayed on a sf object in a ggplot2 graph. Only vertices with a corresponding feature in the sf object are included.

Value

A data frame, produced by fortify, which can be used as data source in ggplot2 graphs.

Examples

net=network::network(matrix(c(0, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0), nrow=4, byrow=TRUE))
network::set.vertex.attribute(net, "name", value=c("a", "b", "c", "d"))
wkb = structure(list("01010000204071000000000000801A064100000000AC5C1641",
"01010000204071000000000000801A084100000000AC5C1441",
"01010000204071000000000000801A044100000000AC5C1241",
"01010000204071000000000000801A024100000000AC5C1841"), class = "WKB")
map=sf::st_sf(id=c("a1", "b2", "c3", "d4"), sf::st_as_sfc(wkb, EWKB=TRUE))
lkptbl=data.frame(id=c("a1", "b2", "c3", "d4"), name=c("a", "b", "c", "d"))
ggnetmap(net, map, lkptbl, "id", "name")

Is data frame a lookup table?

Description

Checks whether a data.frame is a valid lookup table.

Usage

is_lookup_table(lkp, m_name = NULL, n_name = NULL)

Arguments

lkp

A data.frame.

m_name

Optional, a character string with the name of the column in lkp to check against m.

n_name

Optional, a character string with the name of the column in lkp to check against n.

Value

FALSE on error, a vector with m_name and n_name if the lookup table is valid.

Is object a network?

Description

Checks whether an object is a network object or an igraph object, returns message if it's not

Usage

is_network(n)

Arguments

n

Object of class network or igraph.

Value

TRUE if object of class network, FALSE otherwise.

Is object a map?

Description

Checks whether an object is an sf object, returns message if it's not

Usage

is_sf(m)

Arguments

m

Object of class sf.

Value

TRUE if object of classes sf and data.frame, FALSE otherwise.

Description

Checks which vertices of a network object can be represented with features of a sf object.

Usage

link_network_map(m, n, m_name, n_name = "vertex.names")

Arguments

m

Object of class sf.

n

Object of class network or igraph.

m_name

Name of the map field to use for the link.

n_name

Name of the vertex attribute to use for the link, defaults to vertex.names.

Value

On success a list with two vectors, one of features in m present in n, the other of nodes in n present in m, -1 on error.

Description

Checks which vertices of a network object can be represented with features of a sf object with a lookup table.

Usage

link_network_map2(m, n, lkp, m_name = NULL, n_name = NULL)

Arguments

m

Object of class sf.

n

Object of class network or igraph.

lkp

Lookup table, a data.frame.

m_name

Optional character, name of field in m and of column in lkp (first column of lkp is used if NULL).

n_name

Optional character, name of vertex attribute in n and of column in lkp (second column of lkp is used if NULL).

Value

On success a list with two vectors, one of features in m present both in the lookup table and in n, the other of nodes in n present both in the lookup table and in m, -1 on error.

netmap: Plot network and igraph objects on a sf map using ggplot2

Description

The netmap package extends the ggnetwork package by providing functions to plot networks, with vertices usually representing objects with a spatial attribute (cities, regions, countries, users with location data attached etc.) on a map, provided by a sf object (which in turn is able to represent more or less all spatial data available). Networks and maps need not have the same set of elements: if they don't, only the intersection will be represented.

netmap functions

The main function is ggnetmap, which produces a data.frame that is then used as data within ggplot2 calls. For those wishing to use the plot.network or the plot.igraph function to plot the network (without overlaying it on an sf object), both a custom layout function, network.layout.extract_coordinates, and a wrapper that provides convenient manipulation of network and sf objects, netmap_plot, are available.

Plot a network object with a layout based on an sf object

Description

Wrapper for plot.network and plot.igraph using a custom network layout that extracts coordinates of centroids from a sf object. Only vertices with a corresponding feature are plotted.

Usage

netmap_plot(n, m, lkp = NULL, m_name = NULL, n_name = "vertex.names", ...)

Arguments

n

A network or igraph object.

m

A sf object.

lkp

An optional lookup table.

m_name

Optional character, name of field in m and of column in lkp.

n_name

Optional character, name of vertex attribute in n and of column in lkp.

...

Additional parameters passed to plot.network.

Value

A plot of the network.

Examples

net=network::network(matrix(c(0, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0), nrow=4, byrow=TRUE))
network::set.vertex.attribute(net, "name", value=c("a", "b", "c", "d"))
wkb = structure(list("01010000204071000000000000801A064100000000AC5C1641",
"01010000204071000000000000801A084100000000AC5C1441",
"01010000204071000000000000801A044100000000AC5C1241",
"01010000204071000000000000801A024100000000AC5C1841"), class = "WKB")
map=sf::st_sf(id=c("a1", "b2", "c3", "d4"), sf::st_as_sfc(wkb, EWKB=TRUE))
lkptbl=data.frame(id=c("a1", "b2", "c3", "d4"), name=c("a", "b", "c", "d"))
netmap::netmap_plot(net, map, lkptbl, "id", "name")

Layout of a network based on a sf object

Description

Custom layout for plot.network, extracting coordinates of vertices from a sf object. Its result can be used by plot.igraph as well.

Usage

network.layout.extract_coordinates(n, layout.par)

Arguments

n

A network or igraph object. Not used, only for compatibility with plot.network.

layout.par

A list of layout parameters (the only one implemented is layout.par$sf, an sf object whose rows match the order of vertices in n).

Value

A matrix whose rows contain the x,y coordinates of the vertices of n.

Examples

net=network::network(matrix(c(0, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0), nrow=4, byrow=TRUE))
network::set.vertex.attribute(net, "name", value=c("a", "b", "c", "d"))
wkb = structure(list("01010000204071000000000000801A064100000000AC5C1641",
"01010000204071000000000000801A084100000000AC5C1441",
"01010000204071000000000000801A044100000000AC5C1241",
"01010000204071000000000000801A024100000000AC5C1841"), class = "WKB")
map=sf::st_sf(id=c("a1", "b2", "c3", "d4"), sf::st_as_sfc(wkb, EWKB=TRUE))
lkptbl=data.frame(id=c("a1", "b2", "c3", "d4"), name=c("a", "b", "c", "d"))
netmap::network.layout.extract_coordinates(net, list(sf=map))

Reduces network to vertices present on the map

Description

Removes vertices from a network or igraph object which are not present in the link vector produced by link_network_map or link_network_map2.

Usage

reduce_to_map(n, link, n_name)

Arguments

n

A network or igraph object.

link

A vector with the identifiers of the vertices to keep.

n_name

Name of the vertex attribute to filter on.

Value

A network or igraph object with only the vertices listed in link.