partition_tdv_grasp {diffval}

Obtain a partition using a GRASP algorithm

Description

This function obtains a partition of the columns of a given phytosociological matrix, aiming at high values of the Total Differential Value (TDV) using a GRASP algorithm.

Usage

partition_tdv_grasp(m_bin, k, thr = 0.95, verify = TRUE)

Arguments

m_bin

A matrix. A phytosociological table of 0s (absences) and 1s (presences), where rows correspond to taxa and columns correspond to relevés.

k

A numeric giving the number of desired groups.

thr

A numeric giving a threshold value (from 0 to 1 ) with the probability used to compute the sample quantile, in order to get the best m_bin columns from which to select one to be include in the GRASP solution (in each step of the procedure).

verify

A logical. If TRUE (the default) the function verifies if basic features of m_bin data structure are met. Otherwise if FALSE.

Details

This function uses a Greedy Randomized Adaptive Search Procedure (GRASP) to obtain a partition of m_bin. Given a phytosociological table (m_bin, with rows corresponding to taxa and columns corresponding to relevés) this function searches for a k-partition (k, defined by the user) aiming at high values of the TDV. See tdv() for an explanation on the TDV of a phytosociological table.

With thr = 1, the algorithm corresponds to the Greedy algorithm.

Value

A numeric vector, which length is the same as the number of columns of m_bin, with numbers from 1 to k, representing the group to which the respective column was ascribed.

Author(s)

Jorge Orestes Cerdeira and Tiago Monteiro-Henriques. E-mail: [e-mail].

Examples

# Getting the Taxus baccata forests data set
data(taxus_bin)

# Obtaining a partition based on the GRASP algorithm
partition_tdv_grasp(taxus_bin, 3)