Rough Sets and Formal Concept Analysis: Foundations and the Case Studies of Feature Subset Selection and Knowledge Structure Formation

We consider also the paving problem and give a deterministic algorithm to partition a matrix into almost isometric blocks recovering previous results of Bourgain-Tzafriri and

We found that the two variable selection methods had comparable classification accuracy, but that the model selection approach had substantially better accuracy in selecting

The problem of well conditioned column selection that we consider consists of finding the largest subset of columns of X such that the corresponding submatrix has all singular values

Due to the deficiency of the original fuzzy-rough dependency degree in removing redundant features, we proposed a new measure for feature selection based on the

Nevertheless, a number of generic feature construction methods exist: clustering; linear (affine) projections of the original feature space; as well as more sophisti- cated

In more detail, for low values of α, (from 0.005 to 0.11), all the feature sets resulting from the new proposed method lead to better LR performance in com- parison to the feature

knowledge, the algorithm providing the best approximation is the one proposed by Grabisch in [20]. It assumes that in the absence of any information, the most reasonable way of

Construct validity indicates whether the studied operational measures really represent what is investigated according to the research question. The purpose of this study is to