Enabling ad-hoc reuse of private data repositories through schema extraction

Table 1 HCLS core statistics [32] of evaluated datasets, vocabularies and extracted schemata

	HCLS Metric	6.6.1.1	6.6.1.2	6.6.1.3	6.6.1.4	6.6.1.5	6.6.1.6	6.6.1.7
	number of unique	triples	typed entities	subjects	properties	objects	classes	literals
LOV	Vocabulary Corpus	833834	129827	171168	1209	145498	1469	180680
PRs	Dataset	150000	20000	20000	20	10003	3	70717
Vocabulary	schema.org (1)	8427	1617	1619	15	476	31	3193
	foaf (2)	631	84	86	15	38	9	154
	merge of (1), (2)	9058	1701	1705	23	508	38	3335
Schema	directly instantiated	47	23	23	3	5	2	0
	locally inferred	576	95	95	13	71	9	118
	LOV inferred	2345	208	208	87	379	16	850
GenDR	Dataset	11609	1123	1123	24	1232	13	5158
Vocabulary	GenDR Vocabulary	192	20	20	8	6	5	116
Schema	directly instantiated	361	37	37	10	16	7	105
	locally inferred	380	37	37	10	16	7	124
	LOV inferred	911	71	71	58	127	12	370
Orphanet	Dataset	377947	28871	28871	38	42891	29	144773
Vocabulary	Orphanet Vocabulary	402	40	40	9	7	5	239
Schema	directly instantiated	799	67	67	12	41	7	217
	locally inferred	840	68	68	12	41	7	256
	LOV inferred	1380	102	102	59	153	12	506
Homologene	Dataset	7189742	869981	869981	14	1420471	10	2865019
Vocabulary	Homologene Vocabulary	62	7	7	8	6	5	38
Schema	directly instantiated	184	24	24	10	13	7	40
	locally inferred	190	24	24	10	13	7	46
	LOV inferred	721	58	58	58	124	12	292

ISSN: 2041-1480