We present a problem of factor analysis of three-way binary data, i.e. data described by a 3-dimensional binary matrix I, describing a relationship between objects, attributes, and conditions. The problem consists in finding a small number of factors which explain the data. In terms of matrix decompositon, we look for a decomposition of I into three binary matrices, an object-factor matrix A, an attribute-factor matrix B, and a condition-factor matrix C, with the number of factors as small as possible. Compared to other decomposition-based methods, the difference consists in the composition operator and the constraint on A, B, and C to be binary. Due to the space limit, we present the problem statement, a non-technical description of our approach, and, as the main part, an illustrative example.
[1]
Tamara G. Kolda,et al.
Tensor Decompositions and Applications
,
2009,
SIAM Rev..
[2]
Bernhard Ganter,et al.
Formal Concept Analysis: Mathematical Foundations
,
1998
.
[3]
P. Kroonenberg.
Applied Multiway Data Analysis
,
2008
.
[4]
Rudolf Wille,et al.
A Triadic Approach to Formal Concept Analysis
,
1995,
ICCS.
[5]
Vilém Vychodil,et al.
Discovery of optimal factors in binary data via a novel method of matrix decomposition
,
2010,
J. Comput. Syst. Sci..
[6]
Andreas Hotho,et al.
TRIAS--An Algorithm for Mining Iceberg Tri-Lattices
,
2006,
Sixth International Conference on Data Mining (ICDM'06).
[7]
Rudolf Wille,et al.
The Basic Theorem of triadic concept analysis
,
1995
.