ereduce: Find all minimal disjunctive normal forms (DNF) of an input...

In cnaOpt: Optimizing Consistency and Coverage in Configurational Causal Modeling

Find all minimal disjunctive normal forms (DNF) of an input DNF

Description

ereduce builds all minimal disjunctive normal forms corresponding to an input DNF. It is similar to rreduce, which, however, only builds one minimal DNF at random.

Usage

ereduce(cond, x = full.ct(cond), full = !missing(x), 
        simplify2constant = TRUE, maxCombs = 1e7)

Arguments

cond	A character string specifying a disjunctive normal form (DNF); can be either crisp-set or multi-value.
x	A configTable or data.frame; can be either crisp-set or multi-value.
full	Logical; if TRUE, redundancies are eliminated relative to full.ct(x), otherwise relative to x.
simplify2constant	Logical; if TRUE (the default), a tautologous or contradictory cond is reduced to a constant "1" or "0", respectively. If FALSE, a minimal tautology or contradiction, i.e. "A+a" or "A*a", will result.
maxCombs	Maximal number of iterations that will be ran in the most time-consuming step. If the number of necessary iterations exceeds maxCombs, ereduce will stop executing and return an error message stating the necessary number of iterations. Early termination can then be avoided by increasing maxCombs accordingly.

Details

ereduce eliminates conjuncts and disjuncts from a DNF cond as long as the result of condition(cond, x) remains the same. The only required argument is cond. If x is not provided, redundancies are eliminated relative to full.ct(cond).

ereduce generates all redundancy-free forms of cond, while rreduce only returns one randomly chosen one. rreduce is faster than ereduce, but often incomplete. In a nutshell, ereduce searches for minimal hitting sets in cond preventing cond from being false in data x.

Value

A vector of redundancy-free disjunctive normal forms (DNF).

See Also

rreduce, full.ct, conCovOpt, DNFbuild.

Examples

# Logical redundancies.
cond1 <- "A*b + a*B + A*C + B*C"
ereduce(cond1)
rreduce(cond1) # repeated calls generate different outputs
cond2 <- "A*b + a*B + A*B + a*b"
ereduce(cond2)
ereduce(cond2, simplify2constant = FALSE)

# Redundancy elimination relative to simulated cs data.
dat1 <- data.frame(
  A = c(0, 0, 0, 0, 1, 1, 0, 1), 
  B = c(0, 1, 0, 1, 1, 0, 0, 0), 
  C = c(1, 1, 0, 1, 1, 0, 1, 1), 
  D = c(0, 0, 0, 0, 0, 1, 1, 1))
cco1 <- conCovOpt(dat1, "D")
best1 <- selectMax(cco1)
(formula1 <- DNFbuild(best1, outcome = "D", reduce = FALSE))
# ereduce
ereduce(formula1, dat1, full = FALSE)
# rreduce
rreduce(formula1, dat1, full = FALSE)

# Redundancy elimination relative to simulated mv data.
dat2 <- data.frame(
  A = c(3,2,1,1,2,3,2,2,2,1,1,2,3,2,2,2,1,2,3,3,3,1,1,1,3,1,2,1,2,3,3,2,2,2,1,2,2,3,2,1,2,1,3,3),
  B = c(1,2,3,2,1,1,2,1,2,2,3,1,1,1,2,3,1,3,3,3,1,1,3,2,2,1,1,3,3,2,3,1,2,1,2,2,1,1,2,2,3,3,3,3),
  C = c(1,3,3,3,1,1,1,2,2,3,3,1,1,2,2,2,3,1,1,2,1,2,2,3,3,1,2,2,2,3,2,1,1,2,2,2,1,1,1,2,2,1,1,2),
  D = c(3,1,2,2,1,1,1,1,1,1,1,2,2,2,2,2,2,3,3,3,1,1,1,1,1,2,2,2,2,2,3,1,1,1,1,1,2,2,2,2,2,3,3,3),
  E = c(3,2,2,3,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,2,2,2,3,3,3,3,3,3,3,3,3,3,3,3,3)
)
cco2 <- conCovOpt(dat2, "D=3")
best2 <- selectMax(cco2)
(formula2 <- DNFbuild(best2, outcome = "D=3", reduce = FALSE))
# ereduce
ereduce(formula2, dat2, full = FALSE)
# rreduce
rreduce(formula2, dat2, full = FALSE)

# Any Boolean expressions.
cond <- "!(A*B*C)*!(a*b*c)" # or "A + B*!(D + e) <-> C" 
x <- selectCases(cond) 
(cond <- cna:::getCond(x)) # returns a DNF equivalent to cond, but with many redundancies
ereduce(cond)
rreduce(cond)

cnaOpt documentation built on July 8, 2022, 5:06 p.m.