On the Usability of Probably Approximately Correct Implication Bases

We revisit the notion of probably approximately correct implication bases from the literature and present a first formulation in the language of formal concept analysis, with the goal to investigate whether such bases represent a suitable substitute for exact implication bases in practical use cases. To this end, we quantitatively examine the behavior of probably approximately correct implication bases on artificial and real-world data sets and compare their precision and recall with respect to their corresponding exact implication bases. Using a small example, we also provide evidence suggesting that implications from probably approximately correct bases can still represent meaningful knowledge from a given data set.

[1]  Bernhard Ganter,et al.  Conceptual Exploration , 2016, Springer Berlin Heidelberg.

[2]  Tomasz Imielinski,et al.  Mining association rules between sets of items in large databases , 1993, SIGMOD Conference.

[3]  Michael Frazier,et al.  Learning conjunctions of Horn clauses , 2004, Machine Learning.

[4]  Gerd Stumme,et al.  Off to new shores: conceptual knowledge discovery and processing , 2003, Int. J. Hum. Comput. Stud..

[5]  Vincent Duquenne,et al.  Attribute-incremental construction of the canonical implication basis , 2007, Annals of Mathematics and Artificial Intelligence.

[6]  Sergei O. Kuznetsov,et al.  On the Intractability of Computing the Duquenne?Guigues Bas , 2004, J. Univers. Comput. Sci..

[7]  Bart Selman,et al.  Horn Approximations of Empirical Data , 1995, Artif. Intell..

[8]  Francesco Kriegel,et al.  NextClosures: Parallel Computation of the Canonical Base , 2015, CLA.

[9]  V. Duquenne,et al.  Incremental Construction of the Canonical Implication Basis , 2003 .

[10]  Bernhard Ganter,et al.  Two Basic Algorithms in Concept Analysis , 2010, ICFCA.

[11]  Uwe Ryssel,et al.  Fast algorithms for implication bases and attribute exploration using proper premises , 2014, Annals of Mathematics and Artificial Intelligence.

[12]  José L. Balcázar,et al.  Construction and learnability of canonical Horn formulas , 2011, Machine Learning.

[13]  Daniel Borchmann,et al.  Learning terminological knowledge with high confidence from erroneous data , 2014, Joint Workshop of the German Research Training Groups in Computer Science.

[14]  Dana Angluin,et al.  Queries and concept learning , 1988, Machine Learning.

[15]  James B. Nation,et al.  Discovery of the D-basis in binary tables based on hypergraph dualization , 2015, Theor. Comput. Sci..

[16]  Henry Rouanet,et al.  Algèbre linéaire et formalisation de la notion de comparaison , 1968 .

[17]  Vincent Duquenne,et al.  Familles minimales d'implications informatives résultant d'un tableau de données binaires , 1986 .

[18]  Leslie G. Valiant,et al.  A theory of the learnable , 1984, STOC '84.

[19]  Tom Hanika,et al.  Some Experimental Results on Randomly Generating Formal Contexts , 2016, CLA.