First-Order Context and Formal Concept Analysis

Formal Concept Analysis -also called "Galois Lattices"- is an algebraic model based on propositional calculus that is used for symbolic knowledge exploration from a formal context. The aim of this paper is to design the theoretical models required for the extension of Formal Concept Analysis to first-order logic so as to improve both the expression power as a knowledge mining tool upon first order contexts, and the relevance of its results. Our contribution consists in: i) a synthesis of the basic notions of FCA, ii) the design of the Cube Model dedicated to the conjunctions of literals, iii) the design of a complete first-order logic formal concept analysis of first-order contexts. The approach is described from the theoretical point of view, implementations in logic programming and applications are also briefly presented.