Skip to main content
. 2017 Jun 30;18:322. doi: 10.1186/s12859-017-1729-2

Table 1.

Description of the synthetic datasets used in the experiments

Name Attributes Samples Characteristics
CorrAL [24] 99 32 Corr.; F ≫ S
XOR-100 [24] 99 50 N.L; F ≫ S
Parity3+3 [23] 12 64 NL
Monk3 [25] 6 122 No.
SD1 [26] 4020 75 F ≫ S
SD2 [26] 4040 75 F ≫ S
SD3 [26] 4060 75 F ≫ S
Madelon [27] 500 2400 N.L; No.

Corr. stands for correlation, N.L indicates nonlinearity, F ≫ S is used for datasets where the number of features is higher than the number of samples and No. represents noisy data