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 126826221, Article, Gold nugget, page 12 1981-03-07, The Canberra Times (ACT : 1926 - 1995), 77 words, 4 corrections

Raw OCR	Human Corrected	overProof Corrected
Gold nugget	Gold nugget	Gold nugget
MELBOURNE: The world's largest	MELBOURNE: The world's largest	MELBOURNE: The world's largest
gold nugget, the Hand of Faith, found ;	gold nugget, the Hand of Faith, found	gold nugget, the Hand of Faith, found ;
five months ago at Wcdderburri, Vic	five months ago at Wedderburn,	five months ago at Wedderburn, Vic
Victoria, has left Australia.?	Victoria, has left Australia.	Victoria, has left Australia.?
The 27.2-kilogram nugget with, a;	The 27.2-kilogram nugget with a	The 27.2 kilogram nugget with, a;
$ I million price tag, will either be :	$1 million price tag, will either be	$ I million price tag, will either be
displayed in the Golden Nugget Casino V	displayed in the Golden Nugget Casino	displayed in the Golden Nugget Casino V
in Las Vegas, which already dispays the't	in Las Vegas, which already displays the	in Las Vegas, which already dispays the't
Robins nugget valued at $800,000 (dug "	Robins nugget valued at $800,000 (dug	Robins nugget valued at $800,000 (dug "
up in Bcndigo five years ago), or in a r	up in Bendigo five years ago), or in a	up in Bendigo five years ago), or in a r
.new $140 million complex in Atlanta,''	new $140 million complex in Atlanta,	new $140 million complex in Atlanta,''
Georgia. _[	Georgia.	Georgia. _[

Identified overProof corrections	WEDDERBURN BENDIGO
Identified overProof non-corrections	DISPLAYS

	Word count	OCR accuracy %	overProof accuracy %	Errors corrected %
All Words	64	93.8	96.9	50.0
Searchability of unique words	47	93.6	97.9	66.7
Weighted Words		94.5	98.2	66.7

NLA Trial Articles from 1981

Article ID 126826221, Article, Gold nugget, page 12 1981-03-07, The Canberra Times (ACT : 1926 - 1995), 77 words, 4 corrections

Accumulated stats for 1 articles from year 1981