natPsoho: Learn a DBN structure with a PSO approach

In dbnR: Dynamic Bayesian Network Learning and Inference

Learn a DBN structure with a PSO approach

Description

Given a dataset and the desired Markovian order, this function returns a DBN structure ready to be fitted. Original algorithm at https://link.springer.com/chapter/10.1007/978-3-030-86271-8_14

Usage

natPsoho(
  dt,
  size,
  f_dt = NULL,
  n_inds = 50,
  n_it = 50,
  in_cte = 1,
  gb_cte = 0.5,
  lb_cte = 0.5,
  v_probs = c(10, 65, 25),
  r_probs = c(-0.5, 1.5),
  score = "bge",
  p = 0.06,
  cte = TRUE
)

Arguments

dt	a data.table with the data of the network to be trained
size	Maximum number of timeslices of the DBN allowed. Markovian order 1 equals size 2, and so on
f_dt	previously folded dataset, in case some specific rows have to be removed after the folding
n_inds	Number of particles used in the algorithm
n_it	Maximum number of iterations that the algorithm can perform
in_cte	parameter that varies the effect of the inertia
gb_cte	parameter that varies the effect of the global best
lb_cte	parameter that varies the effect of the local best
v_probs	vector that defines the random velocity initialization probabilities
r_probs	vector that defines the range of random variation of gb_cte and lb_cte
score	bnlearn score function used
p	parameter of the truncated geometric distribution for sampling edges
cte	a boolean that determines whether the inertia, global best and local best parameters remain constant or vary as the algorithm progresses. Inertia and local best values decrease as the global best increases, to favor exploration at first and exploitation at the end

Value

A 'dbn' object with the structure of the best network found

dbnR documentation built on June 22, 2024, 10:54 a.m.