Table 10

Time performance (in ms) of classification algorithms on datasets
Diabetes dataset Alzheimer’s dataset Antibodies dataset
p value< 5x10-13 5x10-10 5x10-7 5x10-4 5x10-5 5x10-4 5x10-3 5x10-2 5x10-8 5x10-7 5x10-6 5x10-5
R. Tree 337 408 571 1809 184 200 218 491 250 265 608 1478
KNN 265 333 585 3016 130 156 239 607 187 234 414 910
Hyper Pipes 226 274 630 2486 119 259 423 602 281 312 736 2180
Naïve Bayes 250 456 1120 4780 182 340 500 1158 265 362 892 2480
VFI 299 561 1384 7440 187 337 623 1357 280 368 1379 3000
J48 415 833 3718 16581 166 256 712 1385 468 880 3011 11731
K star 468 1387 4150 25974 187 260 666 2349 299 562 2340 6341
SVM 3313 3635 5304 10496 1054 1108 1389 2722 18297 18372 23712 29009
R. Forest 5717 11889 18254 50087 952 1852 4843 8032 5004 6749 13848 21452
M5P 701 2583 7717 50290 290 524 2324 8563 2632 4711 12033 23452
Bayes Net 718 2087 5653 55672 334 662 4996 9031 733 1140 3394 25000
K means 2618 6651 11876 85955 593 1123 7212 12405 850 908 3442 29658
SLR 11215 26380 79308 632840 1330 3413 22625 48215 17389 20649 89107 605365
LDA 683 1044 7994 658668 402 699 35568 869523 1512 2018 17373 632983
Logistic R. 1204 2592 24687 1589092 629 1651 48659 1146783 1654 9379 255103 1315256
ASC 864 3504 32836 5444533 518 1859 36849 2465021 1217 1763 25496 4565896
MLP 23759 314076 4572305 dnf 2057 30342 2789485 dnf 22916 156905 3277395 dnf

Table showing time performance in milliseconds on all level of significance for three datasets. MLP were among the slowest with dnf: “Did not finish”. Time measurements less than 10 seconds are marked in bold.

Kukreja et al.

Kukreja et al. BMC Bioinformatics 2012 13:139   doi:10.1186/1471-2105-13-139

Open Data