id	sid	eid	entity	type
alusur-6996	1	1	25	CARDINAL
alusur-6996	1	2	2017	DATE
alusur-6996	1	3	1	CARDINAL
alusur-6996	1	4	13 1.1	CARDINAL
alusur-6996	1	5	team1	PERSON
alusur-6996	1	6	the leipzig university	ORG
alusur-6996	1	7	alexander von humboldt	PERSON
alusur-6996	1	8	arabic	LANGUAGE
alusur-6996	1	9	the high nineties	DATE
alusur-6996	2	1	seven	CARDINAL
alusur-6996	2	2	arabic	LANGUAGE
alusur-6996	2	3	over 7,000	CARDINAL
alusur-6996	2	4	87,000	CARDINAL
alusur-6996	2	5	1	CARDINAL
alusur-6996	3	1	actual2	FAC
alusur-6996	3	2	arabic	LANGUAGE
alusur-6996	3	3	benjamin kiessling	PERSON
alusur-6996	4	1	1	CARDINAL
alusur-6996	5	1	sarah bowen	PERSON
alusur-6996	5	2	aga khan university, institute	ORG
alusur-6996	5	3	muslim	NORP
alusur-6996	5	4	london	GPE
alusur-6996	5	5	g. romanov	PERSON
alusur-6996	5	6	leipzig university	ORG
alusur-6996	5	7	june 2017	DATE
alusur-6996	5	8	university of vienna	ORG
alusur-6996	5	9	matthew thomas miller	PERSON
alusur-6996	5	10	roshan institute	ORG
alusur-6996	5	11	persian	NORP
alusur-6996	5	12	university of maryland	ORG
alusur-6996	6	1	benjamin kiessling	PERSON
alusur-6996	6	2	mittagessen@l.unchti.me	GPE
alusur-6996	7	1	2	CARDINAL
alusur-6996	7	2	persian	NORP
alusur-6996	7	3	arabic	LANGUAGE
alusur-6996	7	4	sakhr	GPE
alusur-6996	7	5	abbyy finereader	ORG
alusur-6996	7	6	readiris	PERSON
alusur-6996	8	1	90	CARDINAL
alusur-6996	8	2	99.8%	PERCENT
alusur-6996	9	1	abbyy	ORG
alusur-6996	9	2	readiris	PERSON
alusur-6996	9	3	between 65% and 75%	PERCENT
alusur-6996	10	1	arabic	LANGUAGE
alusur-6996	13	1	benjamin kiessling	PERSON
alusur-6996	13	2	matthew thomas miller	PERSON
alusur-6996	13	3	maxim g. romanov	PERSON
alusur-6996	13	4	sarah bowen savant https://github.com/mittagessen/kraken mailto	PERSON
alusur-6996	13	5	sarah.savant%40aku.edu?subject= mailto	PERSON
alusur-6996	13	6	mtmiller%40umd.edu?subject= mailto:	PERSON
alusur-6996	13	7	2 •	TIME
alusur-6996	13	8	kiessling	GPE
alusur-6996	13	9	miller	GPE
alusur-6996	13	10	romanov, & savant	ORG
alusur-6996	13	11	25	CARDINAL
alusur-6996	13	12	2017	CARDINAL
alusur-6996	13	13	arabic	LANGUAGE
alusur-6996	13	14	persian	NORP
alusur-6996	13	15	arabic	NORP
alusur-6996	14	1	7,000	CARDINAL
alusur-6996	14	2	arabic	LANGUAGE
alusur-6996	14	3	1.2	CARDINAL
alusur-6996	14	4	persian	NORP
alusur-6996	14	5	arabic	LANGUAGE
alusur-6996	15	1	persian	NORP
alusur-6996	15	2	arabic	LANGUAGE
alusur-6996	16	1	arabic	NORP
alusur-6996	16	2	arabic	NORP
alusur-6996	17	1	persian	NORP
alusur-6996	17	2	arabic	LANGUAGE
alusur-6996	18	1	persian	NORP
alusur-6996	18	2	arabic	LANGUAGE
alusur-6996	18	3	decades	DATE
alusur-6996	19	1	4.	CARDINAL
alusur-6996	20	1	5	CARDINAL
alusur-6996	20	2	2	CARDINAL
alusur-6996	21	1	1	CARDINAL
alusur-6996	21	2	al-wusṭā	PERSON
alusur-6996	21	3	25	CARDINAL
alusur-6996	21	4	2017	CARDINAL
alusur-6996	21	5	• 3	TIME
alusur-6996	22	1	ocr	ORG
alusur-6996	23	1	persian	GPE
alusur-6996	23	2	arabic	LANGUAGE
alusur-6996	23	3	kraken	ORG
alusur-6996	23	4	kraken ibn ocropus	PERSON
alusur-6996	23	5	1	CARDINAL
alusur-6996	23	6	benjamin kiessling	PERSON
alusur-6996	23	7	leipzig university’s	ORG
alusur-6996	23	8	alexander von humboldt	PERSON
alusur-6996	24	1	first	ORDINAL
alusur-6996	25	1	arabic	LANGUAGE
alusur-6996	25	2	section 4.1	LAW
alusur-6996	26	1	2.1	CARDINAL
alusur-6996	26	2	model7	GPE
alusur-6996	26	3	1,466	CARDINAL
alusur-6996	26	4	ibn al-faqīh’s	PERSON
alusur-6996	26	5	kitāb al-buldān	PERSON
alusur-6996	27	1	first	ORDINAL
alusur-6996	27	2	6	CARDINAL
alusur-6996	28	1	kraken	ORG
alusur-6996	28	2	kitāb al-ḥashāʾish	PERSON
alusur-6996	28	3	al-ṭibbī	PERSON
alusur-6996	28	4	ub	ORG
alusur-6996	28	5	289	CARDINAL
alusur-6996	31	1	1	CARDINAL
alusur-6996	31	2	approximately 24 hours	TIME
alusur-6996	32	1	8	CARDINAL
alusur-6996	32	2	300	CARDINAL
alusur-6996	36	1	1	CARDINAL
alusur-6996	36	2	4 •	TIME
alusur-6996	36	3	kiessling	GPE
alusur-6996	36	4	miller	GPE
alusur-6996	36	5	romanov, & savant	ORG
alusur-6996	36	6	25	CARDINAL
alusur-6996	36	7	2017	CARDINAL
alusur-6996	37	1	97.56%	PERCENT
alusur-6996	37	2	99.68%	PERCENT
alusur-6996	37	3	arabic	NORP
alusur-6996	38	1	2	CARDINAL
alusur-6996	38	2	ibn al-faqīh’s	PERSON
alusur-6996	38	3	kitāb al-buldān	PERSON
alusur-6996	38	4	six	CARDINAL
alusur-6996	39	1	ibn al-faqīh	PERSON
alusur-6996	39	2	arabic	LANGUAGE
alusur-6996	40	1	at least two	CARDINAL
alusur-6996	40	2	four	CARDINAL
alusur-6996	40	3	two	CARDINAL
alusur-6996	40	4	www.archive.org	ORG
alusur-6996	40	5	2	CARDINAL
alusur-6996	40	6	#1-6	MONEY
alusur-6996	42	1	9	CARDINAL
alusur-6996	43	1	200	CARDINAL
alusur-6996	43	2	3-4	CARDINAL
alusur-6996	44	1	10	CARDINAL
alusur-6996	45	1	200	CARDINAL
alusur-6996	47	1	2	CARDINAL
alusur-6996	47	2	25	CARDINAL
alusur-6996	47	3	2017	CARDINAL
alusur-6996	47	4	• 5	DATE
alusur-6996	47	5	#1-4	MONEY
alusur-6996	48	1	0	MONEY
alusur-6996	48	2	97.61%	PERCENT
alusur-6996	48	3	97.8%	PERCENT
alusur-6996	48	4	arabic	LANGUAGE
alusur-6996	48	5	#1-4	MONEY
alusur-6996	48	6	#5-6	MONEY
alusur-6996	50	1	#1-4	MONEY
alusur-6996	50	2	1	CARDINAL
alusur-6996	51	1	3	CARDINAL
alusur-6996	52	1	2	CARDINAL
alusur-6996	52	2	#1-2	MONEY
alusur-6996	52	3	0	MONEY
alusur-6996	52	4	0	MONEY
alusur-6996	53	1	two	CARDINAL
alusur-6996	53	2	kashīdas	ORG
alusur-6996	53	3	0.3 kashīda	QUANTITY
alusur-6996	53	4	2	CARDINAL
alusur-6996	53	5	0	MONEY
alusur-6996	54	1	3	CARDINAL
alusur-6996	54	2	6 •	CARDINAL
alusur-6996	54	3	kiessling	GPE
alusur-6996	54	4	miller	GPE
alusur-6996	54	5	romanov, & savant	ORG
alusur-6996	54	6	25	CARDINAL
alusur-6996	54	7	2017	CARDINAL
alusur-6996	54	8	3	CARDINAL
alusur-6996	54	9	2	MONEY
alusur-6996	55	1	4	CARDINAL
alusur-6996	55	2	2	MONEY
alusur-6996	56	1	#1-4	MONEY
alusur-6996	57	1	4	CARDINAL
alusur-6996	58	1	2.2	CARDINAL
alusur-6996	58	2	2	MONEY
alusur-6996	59	1	one	CARDINAL
alusur-6996	60	1	approximately 1,000	CARDINAL
alusur-6996	61	1	12-24 hours	TIME
alusur-6996	63	1	2	CARDINAL
alusur-6996	64	1	arabic	LANGUAGE
alusur-6996	67	1	section 3.1	LAW
alusur-6996	68	1	4	CARDINAL
alusur-6996	68	2	25	CARDINAL
alusur-6996	68	3	2017	CARDINAL
alusur-6996	68	4	• 7	DATE
alusur-6996	68	5	kraken	ORG
alusur-6996	68	6	#1-4	MONEY
alusur-6996	69	1	0	MONEY
alusur-6996	69	2	2	CARDINAL
alusur-6996	70	1	#1-4	MONEY
alusur-6996	70	2	second	ORDINAL
alusur-6996	70	3	the high nineties	DATE
alusur-6996	70	4	4	CARDINAL
alusur-6996	71	1	4	MONEY
alusur-6996	71	2	83.42%	PERCENT
alusur-6996	71	3	arabic	LANGUAGE
alusur-6996	71	4	first	ORDINAL
alusur-6996	71	5	0	MONEY
alusur-6996	71	6	99.18%	PERCENT
alusur-6996	72	1	#1-3	MONEY
alusur-6996	72	2	90.90%	PERCENT
alusur-6996	72	3	97.71%	PERCENT
alusur-6996	72	4	87.90%	PERCENT
alusur-6996	72	5	98.47%	PERCENT
alusur-6996	72	6	72.78% to 97.59%	PERCENT
alusur-6996	74	1	arabic	LANGUAGE
alusur-6996	76	1	models12	FAC
alusur-6996	76	2	two	CARDINAL
alusur-6996	77	1	first	ORDINAL
alusur-6996	78	1	secondly	ORDINAL
alusur-6996	78	2	12	CARDINAL
alusur-6996	80	1	2	CARDINAL
alusur-6996	80	2	kraken	ORG
alusur-6996	80	3	8 •	TIME
alusur-6996	80	4	kiessling	GPE
alusur-6996	80	5	miller	GPE
alusur-6996	80	6	romanov, & savant	ORG
alusur-6996	80	7	25	CARDINAL
alusur-6996	80	8	2017	CARDINAL
alusur-6996	80	9	arabic	LANGUAGE
alusur-6996	81	1	arabic	LANGUAGE
alusur-6996	81	2	the last century and a half	DATE
alusur-6996	81	3	arabic	LANGUAGE
alusur-6996	82	1	3.1	CARDINAL
alusur-6996	82	2	two	CARDINAL
alusur-6996	82	3	arabic	LANGUAGE
alusur-6996	82	4	the high nineties	DATE
alusur-6996	83	1	up to 97.5%	PERCENT
alusur-6996	84	1	#1-4	MONEY
alusur-6996	84	2	#1-2	MONEY
alusur-6996	85	1	microtask	NORP
alusur-6996	85	2	3	CARDINAL
alusur-6996	85	3	3	CARDINAL
alusur-6996	85	4	ocr pipeline flowchart	ORG
alusur-6996	85	5	25	CARDINAL
alusur-6996	85	6	2017	CARDINAL
alusur-6996	85	7	• 9	DATE
alusur-6996	87	1	over a dozen	CARDINAL
alusur-6996	88	1	persian	NORP
alusur-6996	88	2	hebrew	LANGUAGE
alusur-6996	88	3	syriac typefaces	ORG
alusur-6996	88	4	spring 2018	DATE
alusur-6996	88	5	persian	NORP
alusur-6996	88	6	arabic	LANGUAGE
alusur-6996	88	7	summer 2018	DATE
alusur-6996	89	1	turkish	NORP
alusur-6996	89	2	urdu	GPE
alusur-6996	89	3	persian	NORP
alusur-6996	89	4	arabic	LANGUAGE
alusur-6996	89	5	the middle east	LOC
alusur-6996	93	1	hundreds	CARDINAL
alusur-6996	93	2	thousands	CARDINAL
alusur-6996	95	1	4.1	CARDINAL
alusur-6996	96	1	benjamin kiessling	PERSON
alusur-6996	96	2	lu	ORG
alusur-6996	96	3	alexander von humboldt	PERSON
alusur-6996	96	4	digital humanities	ORG
alusur-6996	96	5	kraken	ORG
alusur-6996	96	6	library.16 kraken	PERSON
alusur-6996	97	1	13	CARDINAL
alusur-6996	98	1	1,000	CARDINAL
alusur-6996	98	2	turkish	NORP
alusur-6996	98	3	urdu	GPE
alusur-6996	98	4	arabic	LANGUAGE
alusur-6996	98	5	persian	NORP
alusur-6996	99	1	14	CARDINAL
alusur-6996	100	1	15	CARDINAL
alusur-6996	101	1	16	DATE
alusur-6996	102	1	10 •	DATE
alusur-6996	102	2	kiessling	GPE
alusur-6996	102	3	miller	GPE
alusur-6996	102	4	romanov, & savant	ORG
alusur-6996	102	5	25	CARDINAL
alusur-6996	102	6	2017	CARDINAL
alusur-6996	102	7	1997	DATE
alusur-6996	104	1	2	CARDINAL
alusur-6996	106	1	~50	ORG
alusur-6996	106	2	60	CARDINAL
alusur-6996	107	1	2-5%).19	CARDINAL
alusur-6996	109	1	5.1	CARDINAL
alusur-6996	109	2	leipzig university, university of maryland	ORG
alusur-6996	109	3	aga khan university	ORG
alusur-6996	109	4	london	GPE
alusur-6996	110	1	elijah cooke	PERSON
alusur-6996	110	2	roshan institute	ORG
alusur-6996	110	3	samar ata	PERSON
alusur-6996	110	4	mohammad	PERSON
alusur-6996	110	5	mohammad meqdad	PERSON
alusur-6996	111	1	alexander von humboldt	PERSON
alusur-6996	111	2	keshavarz (roshan institute	ORG
alusur-6996	111	3	persian	NORP
alusur-6996	111	4	david taylor	PERSON
alusur-6996	112	1	17	CARDINAL
alusur-6996	113	1	graves et al.	PERSON
alusur-6996	113	2	2006	DATE
alusur-6996	113	3	2013	DATE
alusur-6996	114	1	18	CARDINAL
alusur-6996	117	1	25	CARDINAL
alusur-6996	117	2	2017	CARDINAL
alusur-6996	117	3	• 11	DATE
alusur-6996	117	4	sepp	PERSON
alusur-6996	117	5	jürgen schmidhuber	PERSON
alusur-6996	119	1	9.8	CARDINAL
alusur-6996	119	2	1997	DATE
alusur-6996	119	3	1735-1780	CARDINAL
alusur-6996	120	1	alex	PERSON
alusur-6996	121	1	23rd	ORDINAL
alusur-6996	122	1	acm	ORG
alusur-6996	122	2	2006	DATE
alusur-6996	123	1	breuel	PERSON
alusur-6996	123	2	thomas m.	PERSON
alusur-6996	124	1	english	LANGUAGE
alusur-6996	124	2	fraktur	ORG
alusur-6996	124	3	12th	ORDINAL
alusur-6996	125	1	2013	DATE
alusur-6996	126	1	clstm neural network	ORG
alusur-6996	126	2	https://openphilology.github.io/nidaba/ kraken	PERSON
alusur-6996	126	3	arabic	NORP
alusur-6996	126	4	two	CARDINAL
alusur-6996	126	5	second	ORDINAL
alusur-6996	127	1	ibn al-faqīh	PERSON
alusur-6996	128	1	late 3rd/9th century	DATE
alusur-6996	129	1	kitāb al-buldān	PERSON
alusur-6996	130	1	ed	PERSON
alusur-6996	131	1	yūsuf al-hādī	PERSON
alusur-6996	132	1	beirut	GPE
alusur-6996	132	2	al-kutub	ORG
alusur-6996	132	3	1996	DATE
alusur-6996	133	1	#1	QUANTITY
alusur-6996	133	2	ibn al-athīr	PERSON
alusur-6996	134	1	al-kāmil	PERSON
alusur-6996	134	2	al-taʾrīkh	PERSON
alusur-6996	135	1	ed	PERSON
alusur-6996	135	2	ʿabd allāh	PERSON
alusur-6996	136	1	beirut	PERSON
alusur-6996	136	2	al-kutub	PERSON
alusur-6996	136	3	1415/1994	DATE
alusur-6996	137	1	d. 276/889	PERSON
alusur-6996	138	1	kitāb adab al-kātib	PERSON
alusur-6996	139	1	ed	PERSON
alusur-6996	140	1	muḥammad al-dālī	PERSON
alusur-6996	141	1	al-risāla	ORG
alusur-6996	141	2	n.d.	GPE
alusur-6996	142	1	#3	CARDINAL
alusur-6996	142	2	d.	NORP
alusur-6996	143	1	kitāb al-ḥayawān	PERSON
alusur-6996	144	1	beirut	GPE
alusur-6996	144	2	al-kutub	PERSON
alusur-6996	144	3	1424/2003	DATE
alusur-6996	145	1	fl	ORG
alusur-6996	145	2	second	ORDINAL
alusur-6996	146	1	al-taʾrīkh	PERSON
alusur-6996	147	1	beirut	PERSON
alusur-6996	147	2	n.d.	GPE
alusur-6996	148	1	al-dhahabī	PERSON
alusur-6996	148	2	d. 748/1347	ORG
alusur-6996	150	1	al-maktaba	PERSON
alusur-6996	150	2	al-tawfīqiyya	ORG
alusur-6996	150	3	n.d.	GPE
alusur-6996	151	1	ibn al-jawzī	PERSON
alusur-6996	152	1	al-muntaẓam	PERSON
alusur-6996	153	1	ed	PERSON
alusur-6996	154	1	muḥammad al-qādir	PERSON
alusur-6996	154	2	muṣṭafá al-qādir	PERSON
alusur-6996	155	1	beirut	PERSON
alusur-6996	155	2	al-kutub	PERSON
alusur-6996	155	3	1412/1992	DATE
alusur-6996	156	1	12 •	CARDINAL
alusur-6996	156	2	kiessling	GPE
alusur-6996	156	3	miller	GPE
alusur-6996	156	4	romanov, & savant	ORG
alusur-6996	156	5	25	CARDINAL
alusur-6996	156	6	1	MONEY
alusur-6996	156	7	2	MONEY
alusur-6996	156	8	25	CARDINAL
alusur-6996	156	9	2017	CARDINAL
alusur-6996	156	10	13	CARDINAL
alusur-6996	156	11	4	MONEY
alusur-6996	156	12	3	MONEY
