Notes

Accuracy of OCR and overProof is measured in comparison with the human corrections. We know human corrections in this sample are incomplete, and themselves contain errors, but they are the best we could find automatically from the NLA newspapers corpus, tagged as completely corrected then further filtered to those with at least 3 corrections, at least 40% of lines corrected and lowest third percentage of non-dictionary words.
Accuracy is measured by a separate process from that used to colour words in this output: the colouring process is heuristic, and not completely accurate.
Colour legend:
Text - OCR text corrected by human and/or overProof
Text - human and/or overProof corrections
Text - discrepencies between human and/or overProof
Text - human corrections not applied by overProof
Identified overProof corrections are calculated by the statistical calculation process, and shows those words changed by overProof which ALSO match human corrections. As human corrections are often wrong and incomplete, so too is this list.
Identified overProof non-corrections are calculated by the statistical calculation process, and shows those words in the overProof output which DO NOT MATCH human corrections. As human corrections are often wrong and incomplete, so too is this list. Words marked as [**VANDALISED] are those which have been changed by overProof but not by the human correction; as before, a missed human correction will be (incorrectly) classified as vandalisation by overProof.
Searchability of unique words refers to the distinct words in an article, and how many are present before and after correction. It is measure of how many of the words within an article could be used to find the article using a search engine.
Weighted Words refers to a calculation in which common words count for little (a fraction of a word) and unusual words count for more, in proportion to the log of the inverse of their frequency in the corpus. It may be an indicator of how well distinctive words in an article can be searched before and after correction.

Article ID 625424, Family Notices, Family Notices, page 4 1803-03-05, The Sydney Gazette and New South Wales Advertiser (NSW : 1803 - 1842), 133 words, 34 corrections

Identified overProof corrections	HALF HAWKESBURY LAST ASYLUM CONSUMPTION SHALL ELDEST PRESENT SETTLER CORNWALLIS VEGETABLES KISSING
Identified overProof non-corrections	MORNING WHARF [**VANDALISED] GIRLS

	Word count	OCR accuracy %	overProof accuracy %	Errors corrected %
All Words	106	82.1	94.3	68.4
Searchability of unique words	77	81.8	96.1	78.6
Weighted Words		80.8	96.5	81.6

Article ID 625431, Article, To Correspondents., page 2 1803-03-05, The Sydney Gazette and New South Wales Advertiser (NSW : 1803 - 1842), 115 words, 12 corrections

Identified overProof corrections	CORRESPONDENCE POSSESSED TESTIMONY ISSUING CORRESPONDENTS CLERICAL SOLID SHEEPSHANKS PERSONS COURSE SUCH SUCCEEDS ENSUING /COURT/HOUSE\|COURTHOUSE PROPER
Identified overProof non-corrections	MEREWEATHER PAPER

	Word count	OCR accuracy %	overProof accuracy %	Errors corrected %
All Words	107	81.3	98.1	90.0
Searchability of unique words	77	77.9	97.4	88.2
Weighted Words		74.8	97.5	90.3

Article ID 625438, Article, ADDRESS., page 1 1803-03-05, The Sydney Gazette and New South Wales Advertiser (NSW : 1803 - 1842), 125 words, 18 corrections

Raw OCR	Human Corrected	overProof Corrected
ADD RE'S S,	ADDRESS.	ADD RE'S S,
Innumerable as the Obilacles were 'wliiofe	Innumerable as the Obstacles were which	Innumerable as the Obstacles were 'which
threatened to oppofe our Undertaking, yet	threatened to oppose our Undertaking, yet	threatened to oppose our Undertaking, yet
we are happy to affirm that they were not	we are happy to affirm that they were not	we are happy to affirm that they were not
mfurmountable, however difficult' the talk	insurmountable, however difficult the task	insurmountable, however difficult' the talk
before us.	before us.	before us.
The utility of a PAPER in trreXOLONY,	The utility of a PAPER in the COLONY,	The utility of a PAPER in trreXOLONY,
ar. it mail open a fou/ce of folid information,	as it must open a source of solid information,	ar. it mail open a source of solid information,
will, we hope, be univerfally felt 'and ac-	will, we hope, be universally felt and ac-	will, we hope, be universally felt and acknowledged.
knowledged. We have courted the affiftance	knowledged. We have courted the assistance	We have courted the assistance
of the In g £ *uo u s and Intelligent : -	of the INGENIOUS and INTELLIGENT :—	of the In g s due u s and intelligent : -
We open no channel to Political DifcaiEon, or *	We open no channel to Political Discussion, or	We open no channel to Political DifcaiEon, or *
Perfonal Animad vernon :--Information is	Personal Animadversion :— Information is	Personal Animal vernon :--Information is
our only Putpofe ; that accomplimed, we	our only Purpose ; that accomplished, we	our only Purpose that accomplished, we
'{hill confider that we have doiie our duty, its	shall consider that we have done our duty, in	'shall consider that we have done our duty, its
an exertion to merit the Approbation of the	an exertion to merit the Approbation of the	an exertion to merit the Approbation of the
PUBLIC, and to fecure a liberal Patronage to	PUBLIC, and to secure a liberal Patronage to	PUBLIC, and to secure a liberal Patronage to
the SYDNEY GAZETTE. , .	the SYDNEY GAZETTE.	the SYDNEY GAZETTE. , .

Identified overProof corrections	WHICH DONE OPPOSE SOURCE SOLID SHALL PERSONAL PURPOSE INSURMOUNTABLE UNIVERSALLY ASSISTANCE CONSIDER ACCOMPLISHED SECURE OBSTACLES
Identified overProof non-corrections	INGENIOUS TASK DISCUSSION COLONY MUST ADDRESS ANIMADVERSION

	Word count	OCR accuracy %	overProof accuracy %	Errors corrected %
All Words	107	77.6	91.6	62.5
Searchability of unique words	74	70.3	90.5	68.2
Weighted Words		69.4	90.1	67.7

Article ID 625449, Article, A RUM EFFECT., page 4 1803-03-12, The Sydney Gazette and New South Wales Advertiser (NSW : 1803 - 1842), 347 words, 19 corrections

Raw OCR	Human Corrected	overProof Corrected
A RUM EFFECT.	A RUM EFFECT.	A RUM EFFECT.
" My wife's lo very bad"cry»d Phills.	"My wife's so very bad" cry'd Phill---	" My wife's so very badgered Phillis.
" 1 fear íhe'M never hold it,'*	"I fear she'll never hold it,"	" 1 fear she'll never hold it,"
«* She KEEPS hei bed'» « Mine's worfe"	"She KEEPS her bed" " Mine's worse"	as She KEEPS her beds a Mine's worse"
laid Will	said Will---	said Will
Ci The Jade this morning SOLD it."	"The Jade this morning SOLD it."	Co The Jade this morning SOLD it."
At the commencement of the laft year the	At the commencement of the last year the	At the commencement of the last year the
following dreadful accident happened to the '	following dreadful accident happened to the	following dreadful accident happened to the '
Sir Edward H-nilton rice ihip, bound to	Sir Edward Hamilton rice ship, bound to	Sir Edward Hamilton rice ship, bound to
Bengal; the {hip Sir Edward Hamilton, Cap-	Bengal ; the ship Sir Edward Hamilton, Cap-	Bengal; the ship Sir Edward Hamilton, Captain
tain Robeftfon, in latitude 46. 44. $. and 40.	tain Robertson, in latitude 46. 44. S. and 40.	Robertson, in latitude 46. 44. 9. and 40
E. long, was itruck by lightamg, which car-	E. long. was struck by lightning, which car-	E. long, was struck by lightning, which carried
ried away the fore-top paliant-mait, fhivered	ried away the fore-top gallant-mast, shivered	away the fore-top paliant-mait, shivered
the the fore-top-maft and fore-cap, and went	the fore-top-mast and fore-cap, and went	the the fore-topmast and forehead, and went
through, the deck, abreaft of the fore-maft.	through the deck, abreast of the fore-mast.	through the deck, abreast of the foremast.
It then fpread below, fetting ure to the foUor'I	It then spread below, setting fire to the sailor's	It then spread below, setting fire to the foUor'I
nammocks, wrenching all the iron from their	hammocks, wrenching all the iron from their	hammocks, wrenching all the iron from their
chefts, and taking a diredion aft into the	chests, and taking a direction aft into the	chests, and taking a direction aft into the
cabin ; went off among the arms which were	cabin ; went off among the arms which were	cabin; went off among the arms which were
in that pare of the ftup with a dreadful e»	in that part of the ship with a dreadful ex-	in that part of the flap with a dreadful es
plofion. Captain Robertfon and the crew,	plosion. Captain Robertson and the crew,	portion. Captain Robertson and the crew,
who fortunately were on deck ft rim ç every	who fortunately were on deck furling every	who fortunately were on deck ft rim is every
fail (it blowing a hunicane at the moment),	sail (it blowing a hurricane at the moment),	fail (it blowing a hurricane at the moment),
doug tit the ihip blown up ; but the imoke	thought the ship blown up ; but the smoke	doug fit the ship blown up ; but the smoke
and idlphuroub fmell jiavh g fubfided, tney	and sulphurous smell having subsided, they	and sulphurous smell plays g subsided, they
found ihe was ftili above the waves, and, by	found she was still above the waves, and, by	found the was still above the waves, and, by
timely and fpirked exerriqns, tney weie fuc	timely and spirited exertions, they were suc-	timely and spirited exertions, they were fun
ceibful in ettinguiihing tue flames. The	cessful in extinguishing the flames. The	cessful in extinguishing the flames. The
lightning burnt all the fails in ihe fore-top,	lightning burnt all the sails in the fore-top,	lightning burnt all the fails in the fore-top,
killed ihe bogs, melted the cooking ntenfiis	killed the hogs, melted the cooking utensils	killed the bogs, melted the cooking utensils
in me galley, and knocks uown a J me men	in the galley, and knocked down all the men	in the galley, and knocks down a Jane men
who were m the fore p.u t of me fihip ; but	who were in the fore part of the ship ; but	who were in the for par t of the Ship ; but
none were kiibd, tr-ou^h a nu »ber of the¿n	none were killed, though a number of them	none were killed, through a no other of them
weie blind for foms wii/s. 1 "égale, dunn?	were blind for some days. The gale, during	were blind for foms wires. 1 "gale, during
I which thia alarming ^vcidentn p e^ed, iailci	which this alarming accident happened, lasted	which this alarming accident p eyed, sailed
for lorne uays, and wab uaci itiu(i iiv feveve.	for some days, and was uncommonly severe.	for some days, and was any truth in feveve.
J SYDNEY W; i i K14.-On Saturday Morn-	SYDNEY WHARF.---On Saturday Morn-	J SYDNEY W; i i 14. On Saturday Morn-
ing laft, 5 boa 5 came in irom kiifmg Point	ing last, 5 boats came in from Kissing Point	ing left, 5 boa 5 came in from Kissing Point
with PUR, vegetables, potatoes, poultry, &c.	with fruit, vegetables, potatoes, poultry, &c.	with FUR, vegetables, potatoes, poultry, &c.
Peaches fold from 3d. to 6d. per dozen,	Peaches sold from 3d. to 6d. per dozen,	Peaches sold from 3d. to 6d. per dozen,
melo ia horn 4s. to $s. per J oz. plants 6 L per	melons from 4s. to 5s. per doz. plants 6d. per	more is home 4s. to 8s. per J oz. plants 6 L per
hundied, potatoes from 10s. to 12s. 6d. per	hundred, potatoes from 10s. to 12s. 6d. per	hundred, potatoes from 10s. to 12s. 6d. per
loolbs. : Full grown fowis 3s. each, half	100lbs. : Full grown fowls 3s. each, half	100lbs. Full grown fowls 3s. each, half
j grown du. is. 6. each.	grown do. 1s. 6. each.	j grown du. is. 6. each.

Identified overProof corrections	SULPHUROUS THEY FIRE SHIP SETTING STILL HUNDRED THEM DAYS FOWLS SO SPREAD SMELL SUBSIDED ABREAST KISSING SPIRITED SHIVERED DOWN HAMMOCKS EXERTIONS EXTINGUISHING UTENSILS DURING ROBERTSON HURRICANE SHELL WORSE STRUCK LAST SAID SOME PART DIRECTION GALE HER CHESTS SMOKE
Identified overProof non-corrections	EXPLOSION THOUGHT HOGS PHILL DOZ WHARF SAILS KNOCKED BED [VANDALISED] MELONS FURLING SAILORS DO NUMBER CRYD UNCOMMONLY CAP [VANDALISED] SUCCESSFUL GALLANT HAVING THOUGH BOATS MAST SAIL LASTED SEVERE BAD FRUIT

	Word count	OCR accuracy %	overProof accuracy %	Errors corrected %
All Words	289	66.4	86.2	58.8
Searchability of unique words	174	63.2	83.9	56.2
Weighted Words		63.7	85.1	58.9

NLA Trial Articles from 1803

Article ID 625424, Family Notices, Family Notices, page 4 1803-03-05, The Sydney Gazette and New South Wales Advertiser (NSW : 1803 - 1842), 133 words, 34 corrections

Article ID 625431, Article, To Correspondents., page 2 1803-03-05, The Sydney Gazette and New South Wales Advertiser (NSW : 1803 - 1842), 115 words, 12 corrections

Article ID 625438, Article, ADDRESS., page 1 1803-03-05, The Sydney Gazette and New South Wales Advertiser (NSW : 1803 - 1842), 125 words, 18 corrections

Article ID 625449, Article, A RUM EFFECT., page 4 1803-03-12, The Sydney Gazette and New South Wales Advertiser (NSW : 1803 - 1842), 347 words, 19 corrections

Accumulated stats for 4 articles from year 1803

Raw OCR	Human Corrected	overProof Corrected
DIED	DIED	DIED
At Co;nwallis Farm, on Saturday the 17th	At Cornwallis Farm, on Saturday the 17th	At Cornwallis Farm, on Saturday the 17th
ult. R. Withers, a Mettler of Hawkeibury, of	ult. R. Withers, a Settler of Hawkesbury, of	ult. R. Withers, a Settler of Hawkesbury, of
an apoplexy.	an apoplexy.	an apoplexy.
On Monday lail, the 28 th ult. of a	On Monday last, the 28th ult. of a	On Monday last, the 28th ult. of a
coniumption, Maria Wood. She has left	consumption, Maria Wood. She has left	consumption, Maria Wood. She has left
three children, two of whom are gir!s-~tbs	three children, two of whom are girls---the	three children, two of whom are gir!s-~tbs
eldeil is in the Orphan Afylum.	eldest is in the Orphan Asylum.	eldest is in the Orphan Asylum.
SYDNEY WHARF.-On Saturday Corn-	SYDNEY WHARF.---On Saturday Morn-	SYDNEY WHARFAGE Saturday Corn-
ing lait, between 8" and 9 o'clock, 4 boats	ing last, between 8 and 9 o'clock, 4 boats	ing last, between 8" and 9 o'clock, 4 boats
came in from Kiffing Point with fruit, ve-	came in from Kissing Point with fruit, ve-	came in from Kissing Point with fruit, vegetables.
getables* potatoes, and poultry: Melons 4s,	getables, potatoes, and poultry : Melons 4s.	potatoes, and poultry: Melons 4s,
per dozen, plants 6d. per hundred, potatoes	per dozen, plants 6d. per hundred, potatoes	per dozen, plants 6d. per hundred, potatoes
from 10 to i2s. per soolbs. : Full grown	from 10 to 12s. per 100lbs. : Full grown	from 10 to 12s. per soolbs. : Full grown
fowls fiom 2s. to 2s. 6d. each, halfgrown do,	fowls from 2s. to 2s. 6d. each, half grown do.	fowls from 2s. to 2s. 6d. each, half grown do,
from is. to is. 6d. each.	from 1s. to 1s. 6d. each.	from is to is 6d. each.
*** We ilia.ll hereafter prefent our Readers	We shall hereafter present our Readers	way We shall hereafter present our Readers
with a Table or High Water at Sydney	with a Table of High Water at Sydney	with a Table or High Water at Sydney
and Parramatta.	and Parramatta.	and Parramatta.
Printed at Sydney ; and Adver-	PRINTED AT SYDNEY ; and ADVER-	Printed at Sydney and Advertisements
tisements received by G. Howe."	TISEMENTS RECEIVED BY G. HOWE.	received by G. Howe."
m,		m,

Raw OCR	Human Corrected	overProof Corrected
To Correfpondents.	To Correspondents.	To Correspondents.
Two SLIP BOXES will be put up, in the	Two SLIP BOXES will be put up, in the	Two SLIP BOXES will be put up, in the
courfe of the enfuing Week (one in front	course of the ensuing Week (one in front	course of the ensuing Week (one in front
of the Iifuing Store at Sydney, the other	of the Issuing Store at Sydney, the other	of the Issuing Store at Sydney, the other
in a Window of the Court-houfe, Parra-	in a Window of the Court-house, Parra-	in a Window of the Court-house, Parramatta),
matta), for the Reception oí fuch Articles	matta), for the Reception of such Articles	for the Reception of such Articles
of Information as Perfons who are poíieíTed	of Information as Persons who are possessed	of Information as Persons who are possessed
of the means may think pjoperto contri-	of the means may think proper to contri-	of the means may think proper to contribute.
bute.	bute.
The abundance of animal food through-	The abundance of animal food through-	The abundance of animal food throughout
out the Colony, is an undoubted teftimony	out the Colony, is an undoubted testimony	the Colony, is an undoubted testimony
of the folid advantages that have accrued	of the solid advantages that have accrued	of the solid advantages that have accrued
from a correfpondence with the native in-	from a correspondence with the native in-	from a correspondence with the native inhabitants
habitants of Otaheite.	habitants of Otaheite.	of Otaheite.
By a London reaper we are informed, that	By a London Paper we are informed, that	By a London reaper we are informed, that
jn the weekly account of Uleiical Promo-	in the weekly account of Clerical Promo-	in the weekly account of Clerical Promotions
tions the Rev. Mr. Sheepmanks, fucceeds Dr.	tions the Rev. Mr. Sheepshanks, succeeds Dr.	the Rev. Mr. Sheepshanks, succeeds Dr.
Mere weather in the Rectory of bleating.	Mereweather in the Rectory of Bleating.	Mere weather in the Rectory of bleating.

	Word count	OCR accuracy %	overProof accuracy %	Errors corrected %
All Words	609	73.7	90.6	64.4
Searchability of unique words	402	70.9	90.0	65.8
Weighted Words		69.6	90.1	67.5