R/consensus_similarity.R

In CommKern: Network-Based Communities and Kernel Machine Methods

#' Consensus Similarity
#'
#' Description of the consensus similarity function.
#'
#' This function identifies a single representative partition from a set of
#' partitions that is the most similar to the all others. Here, similarity is
#' taken to be the z-score of the Rand coefficient.
#'
#' @param comm_matrix a matrix whose columns are different partition and whose
#' rows are nodes within a network
#'
#' @return the consensus partition determined by the maximum
#' average pairwise similarity
#'
#' @examples
#'
#' set.seed(7183)
#' x <- sample(x = rep(1:3, 4), 12)
#'
#' y <- sample(x = rep(1:3, 4), 12)
#'
#' z <- sample(x = rep(1:3, 4), 12)
#'
#' xyz_comms_mat <- matrix(c(x,y,z),nrow=length(x),ncol=3)
#' consensus_similarity(xyz_comms_mat)
#'
#' @export
consensus_similarity <- function(comm_matrix) {
    t_comm_matrix <- t(comm_matrix)

    npart <- nrow(t_comm_matrix)

    # Initialize variables
    pairwise_sim <- matrix(NA, nrow = npart, ncol = npart)

    # Calculate the pairwise similarities
    for (i in 1:(npart - 1)) {
        for (j in i:npart) {
            pairwise_sim[i, j] <- zrand(t_comm_matrix[i, ], t_comm_matrix[j, ])
        }
    }

    pairwise_sim <- as.matrix(Matrix::forceSymmetric(pairwise_sim, uplo = "U"))

    # Average pairwise similarity
    average_pairwise_sim <- rowSums(pairwise_sim)/(npart - 1)

    # Extract partition most similar to others
    consensus_part <- t_comm_matrix[which.max(average_pairwise_sim), ]

    return(consensus_part)
}