Package {spatialRegroup}

Title:	Iterative Spatial Regrouping of Administrative Units by Attributive Affinity
Version:	0.1.0
Description:	Evaluates the statistical coherence of existing administrative partitions (e.g. inter-municipal groupings, districts) by identifying spatial units whose attributive profile is more similar to a neighbouring group than to their own. Border units are iteratively reassigned to the group they are most affine with, based on Euclidean or Mahalanobis distance computed on user-supplied numeric variables, with optional per-variable weighting and standardisation. Spatial contiguity is enforced throughout: isolated candidates are reintegrated into their original group, disconnected fragments are resolved, and empty groups are restored. Convergence is monitored via an eta-squared cohesion criterion. The resulting partition can be compared to the original administrative delineation using multilevel models, providing a quantitative measure of boundary inefficiency.
License:	GPL-3
Depends:	R (≥ 4.1.0)
Encoding:	UTF-8
RoxygenNote:	7.3.1
Imports:	dplyr, sf, spdep, igraph, rlang
NeedsCompilation:	no
Packaged:	2026-05-06 08:49:27 UTC; User
Author:	Nicolas Ausello [aut, cre]
Maintainer:	Nicolas Ausello <nicolasausello@yahoo.fr>
Repository:	CRAN
Date/Publication:	2026-05-11 19:10:09 UTC

Rattacher chaque commune candidate au groupe voisin le plus proche

Description

Rattacher chaque commune candidate au groupe voisin le plus proche

Usage

assign_best_group(
  data,
  group_var,
  vars_attr,
  profiles,
  nb = NULL,
  local_profile = TRUE,
  weights = NULL,
  standardize = TRUE
)

Arguments

Value

data.frame avec id_row et best_group

Examples

library(sf)
make_sq <- function(x, y) {
  st_polygon(list(matrix(
    c(x, y, x+1, y, x+1, y+1, x, y+1, x, y),
    ncol = 2, byrow = TRUE
  )))
}
toy <- st_sf(
  EPCI = c('A', 'A', 'B', 'A', 'B', 'B'),
  var1 = c(1.0, 1.2, 3.8, 1.1, 3.9, 3.7),
  var2 = c(2.0, 1.9, 6.0, 2.1, 5.8, 6.1),
  geometry = st_sfc(
    make_sq(0,0), make_sq(1,0), make_sq(2,0),
    make_sq(0,1), make_sq(1,1), make_sq(2,1)
  )
)
toy[['id_row']]    <- seq_len(nrow(toy))
toy[['candidate']] <- c(FALSE, FALSE, TRUE, FALSE, FALSE, FALSE)
nb       <- build_neighbors(toy)
profiles <- compute_profiles(toy, 'EPCI', c('var1', 'var2'))
assign_best_group(toy, 'EPCI', c('var1', 'var2'), profiles, nb = nb)

Construire la matrice de voisinage spatial

Description

Construire la matrice de voisinage spatial

Usage

build_neighbors(data, type = "queen")

Arguments

Value

Un objet nb (spdep)

Examples

library(sf)
make_sq <- function(x, y) {
  st_polygon(list(matrix(
    c(x, y, x+1, y, x+1, y+1, x, y+1, x, y),
    ncol = 2, byrow = TRUE
  )))
}
toy <- st_sf(
  EPCI = c('A', 'A', 'B', 'A', 'B', 'B'),
  var1 = c(1.0, 1.2, 3.8, 1.1, 3.9, 3.7),
  var2 = c(2.0, 1.9, 6.0, 2.1, 5.8, 6.1),
  geometry = st_sfc(
    make_sq(0,0), make_sq(1,0), make_sq(2,0),
    make_sq(0,1), make_sq(1,1), make_sq(2,1)
  )
)
nb <- build_neighbors(toy)
length(nb)  # 6 elements (un par commune)

Calculer les affinites attributaires des communes frontalières

Description

Calculer les affinites attributaires des communes frontalières

Usage

compute_affinity(
  data,
  nb,
  group_var,
  vars_attr,
  method = "euclidean",
  threshold = 0,
  weights = NULL,
  standardize = TRUE
)

Arguments

Value

Un data.frame avec dist_intra, dist_extra, affinite et candidat

Examples

library(sf)
make_sq <- function(x, y) {
  st_polygon(list(matrix(
    c(x, y, x+1, y, x+1, y+1, x, y+1, x, y),
    ncol = 2, byrow = TRUE
  )))
}
toy <- st_sf(
  EPCI = c('A', 'A', 'B', 'A', 'B', 'B'),
  var1 = c(1.0, 1.2, 3.8, 1.1, 3.9, 3.7),
  var2 = c(2.0, 1.9, 6.0, 2.1, 5.8, 6.1),
  geometry = st_sfc(
    make_sq(0,0), make_sq(1,0), make_sq(2,0),
    make_sq(0,1), make_sq(1,1), make_sq(2,1)
  )
)
nb  <- build_neighbors(toy)
aff <- compute_affinity(toy, nb, group_var = 'EPCI',
                        vars_attr = c('var1', 'var2'))
aff[, c('id_row', 'affinite', 'candidat')]

Calculer la coherence interne d'un groupement (eta-carre moyen)

Description

Mesure la part de variance inter-groupe sur la variance totale pour chaque variable attributaire, puis en fait la moyenne. Valeur entre 0 et 1 : plus elle est elevee, plus le groupement est coherent avec les variables. Equivalent a un eta² (rapport de correlation) sans necessiter de modele mixte.

Usage

compute_cohesion(data, group_var, vars_attr)

Arguments

Value

Un scalaire entre 0 et 1 (moyenne des eta² par variable)

Examples

library(sf)
make_sq <- function(x, y) {
  st_polygon(list(matrix(c(x,y,x+1,y,x+1,y+1,x,y+1,x,y),ncol=2,byrow=TRUE)))
}
toy <- st_sf(EPCI=c('A','A','B','A','B','B'),
  var1=c(1.0,1.2,3.8,1.1,3.9,3.7), var2=c(2.0,1.9,6.0,2.1,5.8,6.1),
  geometry=st_sfc(make_sq(0,0),make_sq(1,0),make_sq(2,0),
                  make_sq(0,1),make_sq(1,1),make_sq(2,1)))
compute_cohesion(toy, 'EPCI', c('var1', 'var2'))

Calculer les profils moyens par groupe

Description

Calculer les profils moyens par groupe

Usage

compute_profiles(data, group_var, vars_attr)

Arguments

Value

data.frame des profils moyens par groupe

Examples

library(sf)
make_sq <- function(x, y) {
  st_polygon(list(matrix(
    c(x, y, x+1, y, x+1, y+1, x, y+1, x, y),
    ncol = 2, byrow = TRUE
  )))
}
toy <- st_sf(
  EPCI = c('A', 'A', 'B', 'A', 'B', 'B'),
  var1 = c(1.0, 1.2, 3.8, 1.1, 3.9, 3.7),
  var2 = c(2.0, 1.9, 6.0, 2.1, 5.8, 6.1),
  geometry = st_sfc(
    make_sq(0,0), make_sq(1,0), make_sq(2,0),
    make_sq(0,1), make_sq(1,1), make_sq(2,1)
  )
)
toy[['candidate']] <- FALSE
profiles <- compute_profiles(toy, 'EPCI', c('var1', 'var2'))
profiles

Reintegrer les communes candidates isolees

Description

Reintegrer les communes candidates isolees

Usage

remove_isolates(data, nb)

Arguments

Value

data avec candidats isoles remis a FALSE

Examples

library(sf)
make_sq <- function(x, y) {
  st_polygon(list(matrix(c(x,y,x+1,y,x+1,y+1,x,y+1,x,y), ncol=2, byrow=TRUE)))
}
toy <- st_sf(EPCI=c('A','A','B','A','B','B'),
  var1=c(1.0,1.2,3.8,1.1,3.9,3.7), var2=c(2.0,1.9,6.0,2.1,5.8,6.1),
  geometry=st_sfc(make_sq(0,0),make_sq(1,0),make_sq(2,0),
                  make_sq(0,1),make_sq(1,1),make_sq(2,1)))
toy[['candidate']] <- c(FALSE, TRUE, FALSE, FALSE, FALSE, FALSE)
nb  <- build_neighbors(toy)
res <- remove_isolates(toy, nb)
res[['candidate']]

Regroupement spatial iteratif par affinite attributaire

Description

Regroupement spatial iteratif par affinite attributaire

Usage

spatialRegroup(
  data,
  group_var,
  vars_attr,
  method = "euclidean",
  threshold = 0,
  iterations = 1,
  min_group = 1,
  nb_type = "queen",
  remove_isolates = TRUE,
  weights = NULL,
  standardize = TRUE,
  local_profile = TRUE,
  cohesion_tol = 0.001,
  verbose = TRUE
)

Arguments

Value

Un sf object identique a l'entree avec les colonnes supplementaires : {group_var}_regroup (affectation finale) et candidate (TRUE si l'unite a ete reclassee).

Examples

library(sf)

# Helper : cree un carre unitaire de coin inferieur gauche (x, y)
make_sq <- function(x, y) {
  st_polygon(list(matrix(
    c(x, y, x+1, y, x+1, y+1, x, y+1, x, y),
    ncol = 2, byrow = TRUE
  )))
}

# Grille 3x2 : 6 communes, 2 groupes (A et B)
toy <- st_sf(
  EPCI  = c('A', 'A', 'B', 'A', 'B', 'B'),
  var1  = c(1.0, 1.2, 3.8, 1.1, 3.9, 3.7),
  var2  = c(2.0, 1.9, 6.0, 2.1, 5.8, 6.1),
  geometry = st_sfc(
    make_sq(0,0), make_sq(1,0), make_sq(2,0),
    make_sq(0,1), make_sq(1,1), make_sq(2,1)
  )
)

res <- spatialRegroup(
  data      = toy,
  group_var = 'EPCI',
  vars_attr = c('var1', 'var2'),
  verbose   = FALSE
)
res[, c('EPCI', 'EPCI_regroup', 'candidate')]