Factorizing Three-Way Binary Data with Triadic Formal Concepts

We present a problem of factor analysis of three-way binary data. Such data is described by a 3-dimensional binary matrix I, describing a relationship between objects, attributes, and conditions. The aim is to decompose I into three binary matrices, an object-factor matrix A, an attribute-factor matrix B, and a condition-factor matrix C, with a small number of factors. The difference from the various decomposition-based methods of analysis of three-way data consists in the composition operator and the constraint on A, B, and C to be binary. We present a theoretical analysis of the decompositions and show that optimal factors for such decompositions are provided by triadic concepts developed in formal concept analysis. Moreover, we present an illustrative example, propose a greedy algorithm for computing the decompositions.

[1]  Aristides Gionis,et al.  What is the Dimension of Your Binary Data? , 2006, Sixth International Conference on Data Mining (ICDM'06).

[2]  Bernhard Ganter,et al.  Formal Concept Analysis: Mathematical Foundations , 1998 .

[3]  Petra Perner,et al.  Advances in Data Mining , 2002, Lecture Notes in Computer Science.

[4]  Rudolf Wille,et al.  A Triadic Approach to Formal Concept Analysis , 1995, ICCS.

[5]  Sergei O. Kuznetsov,et al.  Comparing performance of algorithms for generating concept lattices , 2002, J. Exp. Theor. Artif. Intell..

[6]  Clifford Stein,et al.  Introduction to Algorithms, 2nd edition. , 2001 .

[7]  Alexander A. Frolov,et al.  Boolean Factor Analysis by Attractor Neural Network , 2007, IEEE Transactions on Neural Networks.

[8]  Andreas Hotho,et al.  TRIAS--An Algorithm for Mining Iceberg Tri-Lattices , 2006, Sixth International Conference on Data Mining (ICDM'06).

[9]  Tamara G. Kolda,et al.  Tensor Decompositions and Applications , 2009, SIAM Rev..

[10]  Andrzej Cichocki,et al.  Nonnegative Matrix and Tensor Factorization T , 2007 .

[11]  John F. Sowa,et al.  Conceptual Structures: Applications, Implementation and Theory , 1995, Lecture Notes in Computer Science.

[12]  R. Belohlavek,et al.  Optimal decompositions of matrices with grades , 2008, 2008 4th International IEEE Conference Intelligent Systems.

[13]  Rudolf Wille,et al.  The Basic Theorem of triadic concept analysis , 1995 .

[14]  Vilém Vychodil,et al.  Discovery of optimal factors in binary data via a novel method of matrix decomposition , 2010, J. Comput. Syst. Sci..

[15]  Vilém Vychodil,et al.  Factor Analysis of Incidence Data via Novel Decomposition of Matrices , 2009, ICFCA.

[16]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[17]  Pauli Miettinen,et al.  The Discrete Basis Problem , 2008, IEEE Trans. Knowl. Data Eng..

[18]  Ronald L. Rivest,et al.  Introduction to Algorithms , 1990 .

[19]  Johannes Fürnkranz,et al.  Knowledge Discovery in Databases: PKDD 2006, 10th European Conference on Principles and Practice of Knowledge Discovery in Databases, Berlin, Germany, September 18-22, 2006, Proceedings , 2006, PKDD.

[20]  Radim Belohlávek,et al.  Optimal triangular decompositions of matrices with entries from residuated lattices , 2009, Int. J. Approx. Reason..