Singular Value Decomposition (SVD) Analysis of Networks

This function performs Singular Value Decomposition (SVD) on a network adjacency matrix. It computes species importance based on the dominant singular values and returns entropy, rank, and key visualizations.

Usage

calc_svd_entropy_importance(A, threshold_factor = 1e-06)

Arguments

A

A square adjacency matrix where rows represent prey and columns represent predators. The values represent interaction strengths.

Value

A list containing:

Rank

The numerical rank of the matrix after small values are rounded for numerical stability.

Entropy

The entropy of the singular value distribution.

Prey_Importance

A data frame ranking prey species by their contribution to the largest singular value.

Predator_Importance

A data frame ranking predator species by their contribution to the largest singular value.

Singular_Values

A numeric vector of the singular values of the matrix.

Plot_Singular_Values

A ggplot object showing the distribution of singular values.

Plot_Prey_Importance

A ggplot object showing the top prey species contributing to the largest singular value.

Plot_Predator_Importance

A ggplot object showing the top predator species contributing to the largest singular value.

See also

calc_svd_entropy for a lighter version returning only entropy and rank.

Examples

results <- calc_svd_entropy_importance(netData[[1]])
#> Warning: The `<scale>` argument of `guides()` cannot be `FALSE`. Use "none" instead as
#> of ggplot2 3.3.4.
#> ℹ The deprecated feature was likely used in the multiweb package.
#>   Please report the issue at <https://github.com/lsaravia/multiweb/issues>.
print(results$Rank)
#> [1] 21
print(results$Entropy)
#> [1] 0.8862512
print(results$Plot_Singular_Values)

print(results$Plot_Prey_Importance)

print(results$Plot_Predator_Importance)