Faster Concept Analysis


Formal concept analysis (FCA) is a useful tool for data mining, knowledge discovery and automated content organization. This tool is applicable to any setting where objects can be classified according to a set of attributes.

Using algorithms for FCA, one can "lift" this lower level classification to a higher level "conceptual lattice" which captures all association implications in a highly organized, logical and graphically fashion.

A concept lattice can be obtained as the closure system generated from attribute concepts (dually, object concepts). There are two strategies to use this as the basis of an algorithm: (a) forming intersections by joining one attribute concept at a time, and (b) repeatedly forming pairwise intersections starting from the attribute concepts. A straightforward translation of (b) to an algorithm suggests that pairwise intersection be performed among all cumulated concepts. We introduce a simple but efficient, multistage algorithm for constructing concept lattices (MCA). MCA only performs intersections among the newly formed concepts from the previous stage, instead of cumulatively. This way, "the speed of concept set generation is significantly improved, making the proposed algorithm outperform other well-known algorithms for generating concept lattice" (quote from a review report).

For more information, see: