HierarchicalVoronoi: The hierarchical clustering based on Voronoi common edges

In DongDong-Zoez/HCV: What the Package Does (One Line, Title Case)

Description

The hierarchical clustering methods for the constraint of spatial homogeneity.

Usage

HierarchicalVoronoi(constraint_domain, optimization_domain, linkage, iterate)

Arguments

constraint_domain:	'[matrix]': The geometry location of the data.
optimization_domain:	'[matrix]': The spatial feature of the data.
linkage:	'[character]': The valid arguments are "single", "complete", "average", "median", "ward", "centroid" and "weighted".
diss:	'[character]': Set diss = "precomputed" for input optimization_domain as a distance matrix
adjacency:	'[boolean]': Set adjacency = T for input constaraint_domain as an adjacency matrix
dist_method:	'[character]': The distance measurement for pairs of points, default is "euclidean"

Details

The function is used to solve the dual problem. A dual problem is to minimize the cost of the spatial feature domain or the domain of your data attribute, at the meanwhile, we are expect that the cluster of geometry domain or constraint domain will form a compact region, namely, each cluster is bounded by some irregular curves and not overlap for each pairs of clusters. To achieve the goal, one is to consider the spatial homogeneity for spatial data, and HCV algorithm is following the idea of constructing the Delaunay triangulation and Voronoi diagram to make an approach for spatially homogeneity.

Value

A 'hclust' object.

Author(s)

DongDong-Zoez <[e-mail]> University of Taiwan NSYSU.

See Also

'synthetic_data', 'SMI'

Examples

# data <- synthetic_data(5, 4, 0.15, 1000, 2)
# result <- HierarchicalVoronoi(data$geo, data$feat, 'ward.D', 2)
# labels <- cutree(result, 3)

DongDong-Zoez/HCV documentation built on Dec. 17, 2021, 5:29 p.m.