Skip to main content

Table 3 Dataset characteristics for city extractions

From: Network-theoretic information extraction quality assessment in the human trafficking domain

Metric

Set1

Set2

Set3

Set4

Set5

Set6

Set7

Set8

No. of unique extractions

328

1,685

1,489

253

253

1,760

1,563

1,865

No. of extractions per advertisement

1.0

3.4720

17.1507

0.7862

0.7818

3.6858

17.3689

3.4722

No. of ads with no extractions

0

64

65

2,465

2,516

0

0

178

No. of ads with at least 1 extraction

11,530

11,466

11,465

9,065

9,014

11,530

11,530

11,352

Precision

1.0

0.2264

0.0456

1.0

1.0

0.2713

0.0576

0.2264

Recall

1.0

0.7862

0.7818

0.7862

0.7818

1.0

1.0

0.7862

F-score

1.0

0.3516

0.0861

0.8803

0.8775

0.4268

0.1089

0.3516