ConvGeNCode/documentation/paper/Benchmark/Datasets.md

Used datasets

The column "feat." is the number of values for each point in the dataset. The value "points total" is the sum of number of points in the minority set and the number of points in the majority set. The column "minority" contains the numer of points in the minority set. The value "common" gives the sum of number of points in the majority set that are also in the minority set and the number of points in the minority set that are also in the majority set. If the value "common" is not zero then even a perfect classifier will have false positive or false negative predictions.

dataset	feat.	points total	minority	common
folding
$~$ abalone (17 vs. 7, 8, 9, 10)	8	2338	58	0
$~$ abalone (9-18)	8	731	42	0
$~$ car good	6	1728	69	0
$~$ car vgood	6	1728	65	0
$~$ flare-F	11	1066	43	93
$~$ hypothyroid	25	3163	151	2
$~$ kddcup (guess passwd vs. satan)	38	1642	53	0
$~$ kr-vs-k (3 vs. 11)	6	2935	81	0
$~$ kr-vs-k (zero-one vs. draw)	6	2901	105	0
$~$ shuttle (2 vs. 5)	9	3316	49	0
$~$ winequality (red 4)	11	1599	53	0
$~$ yeast4	10	1484	51	0
$~$ yeast5	10	1484	44	0
$~$ yeast6	10	1484	35	0
imblearn
$~$ webpage	300	34780	981	152
$~$ mammography	6	11183	260	3329
$~$ protein homo	74	145751	1296	0
$~$ ozone level	72	2536	73	0
kaggle
$~$ creditcard	30	284807	492	0