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

Applied Network Science

Table 9 Single-point structural network metrics for City extraction datasets

Metric	Set1	Set2	Set3	Set4	Set5	Set6	Set7	Set8
Order	11492	11460	11465	9039	8988	11525	11530	11352
Power-law exponent	0.6431	2.2155	1.3639	0.5924	0.5889	2.3504	1.3709	3.1598
Num. connected components	290	11	2	227	227	7	2	1
Clustering coefficient	0.9944	0.8387	0.8377	0.9947	0.9947	0.8367	0.8367	0.5493
Degree correlation	1.0	0.5641	1.0	1.0	0.5542	0.5542	0.6397	0.6275
Order (of largest connected component)	533	11430	11463	533	513	11510	11528	11352
Algebraic connectivity	532.9999	0.8757	0.4989	533.0	513.0	0.7767	FAIL	7.8806
Vertex connectivity	532	1	3	532	512	1	FAIL	8
Edge connectivity	532	1	3	532	512	1	FAIL	8
Diameter	1	7	5	1	1	7	FAIL	4
Avg. shortest path length	1	3.4945	1.9510	1	1	2.6403	FAIL	2.3603

Since the network is disconnected, all metrics listed below Order (of Largest Connected Component) are computed over the largest connected component. FAIL indicates that the metric could not be computed for that set, usually due to a program timeout or memory issue