%0 Conference Paper
%1 cerf2008datapeeler
%A Cerf, Loïc
%A Besson, Jérémy
%A Robardet, Céline
%A Boulicaut, Jean-Francois
%B Proc. SIAM International Conference on Data Mining SDM'08
%D 2008
%K ol_web2.0 triadic trias three mode fca
%P 37--48
%T Data-Peeler: Constraint-based Closed Pattern Mining in n-ary Relations
%U http://www.siam.org/proceedings/datamining/2008/dm08_04_Cerf.pdf
%X Set pattern discovery from binary relations has been extensively studied during the last decade. In particular, many complete and efficient algorithms which extract frequent closed sets are now available. Generalizing such a task to n-ary relations (n ≥ 2) appears as a timely challenge. It may be important for many applications, e.g., when adding the time dimension to the popular objects × features binary case. The generality of the task — no assumption being made on the relation arity or on the size of its attribute domains — makes it computationally challenging. We introduce an algorithm called Data-Peeler. From a n-ary relation, it extracts all closed n-sets satisfying given piecewise (anti)-monotonic constraints. This new class of constraints generalizes both monotonic and anti-monotonic constraints. Considering the special case of ternary relations, Data-Peeler outperforms the state-of-the-art algorithms CubeMiner and Trias by orders of magnitude. These good performances must be granted to a new clever enumeration strategy allowing an efficient closeness checking. An original application on a real-life 4-ary relation is used to assess the relevancy of closed n-sets constraint-based mining.