SDKL-Miner procedure

Author(s):

Milan Šimůnek, Jan Rauch

Responsibility:

Jan Rauch (theory), Milan Šimůnek (software), Martin Kejkula (help)

Description:

Data mining procedure SDKL-Miner mines for patterns of the form R ~ C / (α, β, Cond). Here R and C are categorial attributes, the attribute R has categories r₁, …, r_K, the attribute C has categories c₁, …, c_L. Further, α, β and Cond are Boolean attributes.

The procedure deals with data matrices. The attributes R and C correspond to columns of the analysed data matrix. Boolean attributes α, β and Cond are derived from the other columns of the data matrix.

The intuitive meaning of the pattern R ~ C / (α, β, Cond) is that the relation of attributes R and C in set a differs from the relation of attributes R and C in set β when the condition given by the Boolean attribute Cond is satisfied.

The symbol ~ is called SDKL-quantifier. It corresponds to a condition imposed on two contingency tables of R and C. The pattern R ~ C / (α, β, Cond) is verified such that this condition is applied on the contingency tables of R and C in data matrices M/α/Cond and M/β/Cond.

Here M is the analysed data matrix and M/α is a data matrix consisting of all rows of data matrix M given by the Boolean attribute α. Further, M/α/Cond is a data matrix consisting of all rows of M/ α satisfying the condition Cond. Let us remark that M/α/Cond can be understood as M/(α ∧ Cond). Analogously for M/β/Cond.

Files to download:

LISp-Miner.Core.zip	25.44 MB	May 18, 2013
All the core files of the LISp-Miner system (LMAdmin, LMDataSource, GUHA procedures, LMTaskPooler, LMSwbImporter/Exporter)

History:

The procedure SDKL-Miner was suggested by J. Rauch in 2003. Reason was the necessity to mine for patterns of the form R ~ C / (α, β, Cond). The second reason was the possibility to use the software tools for dealing with strings of bits developed for the 4ft-Miner procedure. The suggestion was published in [RS 04].

The first version of the procedure with set of simple SDKL-quantifiers was implemented by M. Šimůnek.