Skip to main content

Table 3 Cohen kappa coefficients for all combinations of methods using the approach given above

From: Aggregation of experts: an application in the field of “interactomics” (detection of interactions on the basis of genomic data)

Cohen k

1 region

2 regions

3 regions

4 regions

5 regions

KNN-MDR

0.587

0.537

0.518

0.515

0.511

KNN-BOOST

0.485

0.459

0.502

0.493

0.486

KNN-AES

0.244

0.230

0.246

0.247

0.245

KNN-BHIT

0.177

0.186

0.192

0.200

0.203

KNN-MSH

0.179

0.159

0.152

0.147

0.156

MDR-BOOST

0.608

0.512

0.497

0.507

0.469

MDR-AES

0.354

0.357

0.353

0.349

0.345

MDR-BHIT

0.238

0.255

0.263

0.270

0.275

MDR-MSH

0.208

0.174

0.156

0.155

0.170

BOOST-AES

0.343

0.287

0.277

0.287

0.283

BOOST-BHIT

0.319

0.308

0.287

0.298

0.299

BOOST-MSH

0.195

0.166

0.161

0.158

0.162

AES-BHIT

0.443

0.418

0.406

0.388

0.386

AES-MSH

0.468

0.454

0.415

0.415

0.431

BHIT-MSH

0.312

0.301

0.302

0.308

0.290

  1. These coefficients are estimators based on the set of 500 simulations used to create Figs. 7 and 8