Supriya Ghosh (Editor)

Implication (information science)

Updated on
Edit
Like
Comment
Share on FacebookTweet on TwitterShare on LinkedInShare on Reddit

In formal concept analysis (FCA) and closure system attributes are analyzed via their (simultaneous) occurrences (= objects). The implicational analysis deals with algorithms to find a minimal set of rules, by which attributes can be inferred from others (implications).

Contents

Definition

A pair of sets ( U , V ) is an implication. U A and V A . U is the premise and V is the conclusion. A stands for attributes.

A set W respects U V , if U W  or  V W .

Implications in a Concept Lattice

U V holds in a concept lattice, if every intent respects it. This is true if

  1. iff U V .
  1. iff V U .

An implication U V follows semantically from a set L of implications, if each subset of A respecting L also respects U V . L is closed, if it contains all implications that follow from L.

A set L of implications that hold in a context is called sound. It is called complete if every implications that holds in the context follows from L. L is non-redundant if implications that follow from L are not in L.

If L is a set of implications, then { X A | X respects L } is a closure system.

Implication Basis

Let X X be a closure operator then P is pseudo-closed if

  1. P P
  2. Q P Q P

The set { P P | P pseudo-closed } is sound, complete and non-redundant. It is called Duquenne-Guiges-Basis (D-G-basis) or Stem basis.

There cannot be a implication basis with less implications than the Duquenne-Guiges-Basis, but one can choose implications that are simpler regarding | U | | V | . Such implications are U V with V and U V = .

References

Implication (information science) Wikipedia


Similar Topics