Skip to main content

Table 10 Time performance (in ms) of classification algorithms on datasets

From: Comparative study of classification algorithms for immunosignaturing data

 

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

  1. 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.