The Journal of Open Source Software · 2023

chessboard: An R package for creating networkconnections based on chess moves

Casajus N., Borges E.R., Tabacchi E., Fried G., Mouquet N.

doi.org/10.21105/joss.05753
Download PDF
Scroll down for preview

Key Message

Here we introduce chessboard, an open-source R package (R Core Team (2023)), dedicated to facilitating the process of creating connectivity matrices for samplings designed as regular grids (i.e., the distance between each pair of nodes along one axis is always the same).

Chessboard identifies neighbors based only on node position on a two-dimension non spatial referential.

It can handle spatial networks, but it does not explicitly use geographical coordinates to find neighbors (it is not based on spatial distance). chessboard can handle directed (asymmetric) and undirected (symmetric) spatial (or non-spatial) networks. chessboard offers various methods to detect neighbors, all based on the chess game, allowing the creation of complex connectivity scenarios.

Figure from Casajus et al. 2023
Overview of methods available in chessboard to detect neighbors. Red dots locate the node of interest (5-5) and black dots correspond to detected neighbors. Each column corresponds to a specific method derived from the chess game. Each row illustrates the use of one argument (row 1: default settings; row 2: use of the argument degree; row 3: use of the argument directed; row 4: use of the argument reverse).
Full Article (PDF)