Table 2 Number of samples per client for synthetic data and real journal entry data under i.i.d. and non-i.i.d. settings.

From: Anomaly detection in double-entry bookkeeping data by federated learning system with non-model sharing approach

Dataset

1

2

3

4

5

6

7

8

 

Synthetic data

i.i.d

200

200

200

200

200

200

200

200

non-i.i.d

272

244

162

261

119

289

125

128

Journal entry data

i.i.d

12,780

12,780

12,780

12,780

12,780

12,780

12,780

12,779

non-i.i.d

17,070

22,978

10,708

11,230

14,984

8,525

8,133

8,611

Back to article page