fltr {multiplex}

Principal Filters

Description

A function to find principal filters in a partial order

Usage

fltr(x, PO, ideal = FALSE, rclos)

Arguments

x

the reference element in the partial order (integer or character)

PO

the partial order

ideal

(logical) whether or not the “filter” is an ideal

rclos

(optional and logical) apply reflexive closure?

Details

This function helps to find principal filters or principal ideals for an element in a partial order structure. Such inputs are normally a concept or an object or attribute in the concept together with the associated partial ordering structure of the concepts, which results from Galois derivations. Typically, if the reference element refers to a concept, then it is given as a positive integer indicating the concept label. Another option is to refer to an object or an attribute by a character name, which should be part of the labels of the dimensions of the partial order table with reduced labelling. Principal filters with full labelling are not allowed if the reference element is an object or an attribute. Use an integer for the concept instead.

Value

A named list with the elements in the upset or downset of the principal filter or ideal corresponding to the reference element in the partial order.

Author(s)

Antonio Rivero Ostoic

References

Ganter, B. and R. Wille Formal Concept Analysis – Mathematical Foundations. Springer. 1996.

See Also

galois, partial.order, diagram.

Examples

## Create a data frame
dfr <- data.frame(x=1:3, y=5:7)

## Partial ordering of concepts
PO <- partial.order(galois(dfr),"galois")

## Filter for the first element
fltr(1, PO, rclos=TRUE)