increase_order_complete_search: Determine the (k + 1)'th order t-cherry tree with the highest...

In nvihrs14/tcherry: Learning the structure of tcherry trees

Description

Determine the structure of the (k + 1)'th order t-cherry tree with the highest weight from a k'th order t-cherry tree based on a complete search.

Usage

increase_order_complete_search(tch_cliq, data, ...)

Arguments

tch_cliq	A list containing the cliques of a k'th order t-cherry tree.
data	The data the structure of the tree should be based on.
...	Additional arguments passed to weight_junction_tree.

Details

The algorithm for constructing the (k + 1)'th order t-cherry tree from a k'th order t-cherry tree is based on an atempt to minimize the Kullback-Leibler divergence, by mazimising the weight. All possible structures are determined and the one with the highest weight is chosen.

Note that this procedure is highly inefficient, and only suited for small problems.

Value

A list containing the following components:

• model A list containing the following components:

  • weight The weight of the final (k + 1)'th order t-cherry tree.

  • cliques A list containing the cliques of the (k + 1)'th order t-cherry tree.

  • separators A list containing the separators of a junction tree for the (k + 1)'th order t-cherry tree.

  • adj_matrix The adjacency matrix for the (k + 1)'th order t-cherry tree.

  • n_edges The number of edges in the resulting graph.

• n_models The number of considered models.

Author(s)

Katrine Kirkeby, [e-mail]

Maria Knudsen, [e-mail]

Ninna Vihrs, [e-mail]

See Also

weight_junction_tree for calculation of the weight and increase_order2 for a more efficient, but greedy algorithm.

Examples

set.seed(43)
var1 <- c(sample(c(1, 2), 100, replace = TRUE))
var2 <- var1 + c(sample(c(1, 2), 100, replace = TRUE))
var3 <- var1 + c(sample(c(0, 1), 100, replace = TRUE,
                        prob = c(0.9, 0.1)))
var4 <- c(sample(c(1, 2), 100, replace = TRUE))
var5 <- var2 + var3
var6 <- var1 - var4 + c(sample(c(1, 2), 100, replace = TRUE))
var7 <- c(sample(c(1, 2), 100, replace = TRUE))

data <- data.frame("var1" = as.character(var1),
                   "var2" = as.character(var2),
                   "var3" = as.character(var3),
                   "var4" = as.character(var4),
                   "var5" = as.character(var5),
                   "var6" = as.character(var6),
                   "var7" = as.character(var7))

ChowLiu_cliques <- list(c("var1", "var5"),
                        c("var2", "var5"),
                        c("var3", "var5"),
                        c("var3", "var7"),
                        c("var4", "var6"),
                        c("var5", "var6"))

(tch <- increase_order_complete_search(ChowLiu_cliques, data,
                                       smooth = 0.1))

nvihrs14/tcherry documentation built on Aug. 1, 2020, 6:25 p.m.