Skip to main content

Table 1 Comparison of different annotation methods in terms of AUC

From: Learning Sparse Representations for Fruit-Fly Gene Expression Pattern Image Annotation and Retrieval

Stage range

Number of terms

SVMSpatial+Sparse

SVMSparse

SVMSpatial

SVMGlobal

4-6

10

.8284 ± .0321

.8250 ± .0319

.8064 ± .0321

.7984 ± .0320

 

20

.8310 ± .0286

.8240 ± .0293

.8046 ± .0292

.7965 ± .0302

 

30

.7982 ± .0408

.7892 ± .0399

.7777 ± .0400

.7635 ± .0405

7-8

10

.7808 ± .0285

.7685 ± .0297

.7567 ± .0301

.7472 ± .0293

 

20

.7734 ± .0431

.7619 ± .0427

.7444 ± .0496

.7309 ± .0484

9-10

10

.7917 ± .0260

.7816 ± .0270

.7652 ± .0264

.7538 ± .0265

 

20

.7971 ± .0335

.7829 ± .0344

.7706 ± .0344

.7476 ± .0349

11-12

10

.8526 ± .0248

.8478 ± .0249

.8316 ± .0243

.8257 ± .0240

 

20

.8574 ± .0206

.8437 ± .0214

.8275 ± .0215

.8091 ± .0228

 

30

.8275 ± .0252

.8085 ± .0254

.7940 ± .0274

.7673 ± .0268

 

40

.8193 ± .0290

.7991 ± .0306

.7810 ± .0304

.7560 ± .0321

 

50

.8084 ± .0351

.7894 ± .0363

.7648 ± .0370

.7426 ± .0382

13-16

10

.8807 ± .0221

.8659 ± .0223

.8632 ± .0218

.8398 ± .0225

 

20

.8504 ± .0172

.8301 ± .0182

.8304 ± .0180

.8001 ± .0177

 

30

.8344 ± .0197

.8089 ± .0198

.8066 ± .0190

.7713 ± .0198

 

40

.8175 ± .0196

.7892 ± .0208

.7847 ± .0211

.7496 ± .0223

 

50

.8038 ± .0249

.7748 ± .0208

.7672 ± .0261

.7340 ± .0271

 

60

.7947 ± .0282

.7657 ± .0299

.7613 ± .0300

.7281 ± .0310