Full text of "Analysis of Abbyy OCR Accuracy"

See other formats

Analysis of Abbyy 8 accuracy, normal vs. -fast

Hank Bromley

29 November2007 (revised 3 December2007 to include a small numberof additional errors)

In order to get an approximate sense of how much accuracy is lost in shifting from normal mode to '-fast' mode in Abbyy 8, 1 ran a small sample
of pages under both modes and compared the results. The samples came from three different books, and the results varied significantly, both
across books and across pages within a given book. (In fact, the first page I ran showed better results under '-fast'.) Given the variability, and the
small sample size, these results should be seen as only a rough indicator of what to expect. In particular, I didn't include any books with especially
small print or unusual typefaces, nor any non-English books, so it's hard to predict what will happen with those.

The results are summarized in the table below. On the average, '-fast' mode produced just about twice as many errors as normal mode. In

general, the error rates are fairly low (1 out of every 350-700 characters), with some pages nearly perfect.

Following the summary table are the actual OCR results, shown side-by-side, with normal on the left and '-fast' on the right. Each mode read
correctly some characters that the other mode missed. Where the two differ, the mistaken results are shown in red and the correct results in blue.
Errors made by both are shown in purple. (Other misspellings are not marked at all because they were actually in the originals, and read
"correctly" in both modes.)

A separate document has reduced versions of the actual page images. I thought putting those in a separate document would make it easier to view
the images and OCR results simultaneously.

SUMMARY OF Results

leaf errors ir

errors in

Notes

normal

ocr

-last

ocr

d e p a rtme nta lb u HOOno rtuoft

0004

0005

0006

the only error in -fast was misreading a double quote as two single quotes

newva rietyofc ret3716ric h

0009

0012

these were nearly all repeats of a single error (misreading the abbreviation "pi." as "pi.")

JewishproblemhowOObranrich

0006

this was the only book for which -fast was significantly worse; there was quite a bit of

0009

broken type on the left side of the pages, to which -fast may be less resilient

totals

errors per page

3.0

5.9

errors per 1000 chars

1.4

2.7

15,000 characters total, or -2000 per page

Abbyy 8 accuracy — page 2

department I bullOOrortuoft, leaf

• 1 •*

The U n i ver s i t y

The University

OF NORTH DAKOTA

ESTABLISHED IN EIGHTEEN HUNDRED AND EIGHTY-THREE ESTABLISHED IN EIGHTEEN HUNDRED AND EIGHTY-THREE

FRANK L. McVEY, Ph. [J, LL. Q., President
Grand Forks Bismarck Hebron
University Qevils Lake Minot

1 . T^e College of Liberal Arts offers to men and wo-
men programs of study leading to the degree of
Bachelor of Arts which may be begun in September
or February.

2. The School of Education perpares for the profes-
sion of teaching in secondary and higher schools.
Its graduates receive the degree of Bachelor of Arts
and the Bachelor's Diploma in Teaching. The Mod-
el High School is maintained by the School of
Education as a place of observation and practice.

3. The College of Law offers a three-year course and
grants the degree of Bachelor of Laws.

4. Courses of Study leading to degrees of Mining
Engineer, Electrical Engineer, Mechanical Engi-
neer and Civil Engineer are offered in the School
of Mines and the College of Mechanical and Elec-
trical Engineering.

FRANK L. McVEY, Ph. [J, LL- Q., President
Grand Forks Bismarck Hebron
Universitffl ^evils Lake Minot

1 . T^e College of Liberal Arts offers to men and wo-
men programs of study leading to the degree of
Bachelor of Arts which may be begun in September
or February.

3. The College of Law offers a three-year course and
grants the degree of Bachelor of Laws.

5. The School of ^edicine provides instruction of high

5. The School of ^edlcine provides instruction of high

Abbyy 8 accuracy — page 3

order for two years in medicine based upon two
years of college work. A certificate in medicine
is granted with the A. B. degree.

6. The Graduate Department presents advanced
courses of study leading to the degree of Master
of Arts.

7. The Summer Session provides college and elemen-
tary courses for students and teachers.

8. Extension Lectures and Courses of Study are offer-
ed by the University for persons otherwise unable
to receive academic training.

9. Laboratories and Stations are maintained at Univer-
sity, Devils Lake, Bismarck, Minot and Hebron,
North Dakota.

Information regarding colleges and departments may

be obtained by addressing the Registrar of the

University| University, North Dakota.

University] University, North Dakota.

departmen

talbullOOnortuoft leaf 0005

*m*m ■■«^ %*■■■ ^^^^m m^r ■ *b%p*B^^ ■ «• ■^■^ %»<■ ■ ^^^^^^*^

UNIVERSITY OF NORTH DAKOTA
DEPARTMENTAL BULLETINS

Library February, 1917 No. Q

Library February, 1 9 1 7 No. Q

STATE- WIDE USE OF THE UNIVERSITY
LIBRARY

Abbyy 8 accuracy — page 4

Clarence Wesley Summer, B.A.,
Librarian, University of North Dakota

boys and girls such a love for books that they may be led
into the habit of good reading. We believe that every com-
munity owes it to itself not only to have a library but to
have a librar^ that is in truth a living, vital force in the
community. Much has been said concerning the library of
yesterday, the librarian of which was seemingly content with
collecting and preser^ng books. Librarians of today sti^

believe it to be their duty and privilege to be constantly
gathering into libraries good books and many valuable sources
of information, but not that they may be locked up in cases
and carefully guarded ; not even that they may be ready
for service and usefulness in time of need, but that they
may be, in so far as possible and practicable, sent into the
homes and put into service.

Clarence Wesley Summer, B.A.,
Librarian, University of North Dakota

The spirit of modern library development becomes more
and more truly a spirit of large and far-reaching service;
a spirit of sincerity in attempting to meet some of the real
needs of men and women and to instill into the hearts oQ
boys and girls such a love for books that they may be led
into the habit of good reading. We believe that every com-
munity owes it to itself not only to have a library but to
have a librar^ that is in truth a living, vital force in the
community. Much has been said concerning the library of
yesterday, the librarian of which was seemingly content with
collecting and preser^ng books. Librarians of today stiQ

believe it to be their duty and privilege to be constantly
gathering into libraries good books and many valuable sources

of information, but not that they may be locked up in cases
and carefully guarded ; not even that they may be ready
for service and usefulness in time of need, but that they
may be, in so far as possible and practicable, sent into the
homes and put into service.

department I bullOOrortuoft, leaf

• • • ;

Librarians are coring to see more clearly the possibilities
of their profession and to look upon their work as some-
thing more than routine, essential as routine is in every
library. They are coming to realize that there is truth in
the statement that the real test of the value of the library is

in its use, that the truly useful library is far more than a
building furnished with book- shelves and stocked with row

after row of books, that it is in deed and in truth, as some
one has said, "an active, potential force in the community,
which reaches out and touches and quickens the lives of
individuals in the community, developing and enriching those
lives in every possible way. |

Librarians are coding to see more clearly the possibilities
of their profession and to look upon their work as some-
thing more than routine, essential as routine is in every
library. They are coming to realize that there is truth in
the statement that the real test of the value of the library is

in its use, that the truly useful library is far more than a
building furnished with book- shelves and stocked with row

■■

Abbyy 8 accuracy — page 5

Such is the spirit of present-day library development. No
one can deny, however, if we are to judge from work ac-
complished and lines of service extended, that the spirit
of modern library development would seem to pertain more
to the public library than to that of the college and univer-
sity. The development of the latter, all librarians know,
has not kept pace with that of the former. It has been
too much the tendency of such libraries to serve only their
own immediate communities the student bodies and the
faculties. This is due, no doubt in a large mesure, to the
entirely inadequate support that it receives. It would seem
that university authorities have sometimes failed to recog-
nize the real function of the university library, the larger
scope and function of its activities, the multiplicity of detail
connected with its administration. They have failed to
see that the library is really an institution in itself, with
great possibilities for state-wide service and that, as such,
it should receive liberal financial support; and we have the
conviction that the fault lies partly with the librarians in
not making the influence of their libraries felt more in the
state at large in not putting them upon a plane which
will command the respect which justly belongs to them.

Granting that the first function of a university library
is to meet the demands made upon it by its students and
faculty, is there not another important field of service, lying
at its very door, which could be developed if sufficient sup-
port were given to it? We refer to a state- wide service
in the matter of providing a source of general information
and dissemination of knowledge that would benefit the en-

Such is the spirit of present-day library development. No
one can deny, however, if we are to judge from work ac-
complished and lines of service extended, that the spirit
of modern library development would seem to pertain more
to the public library than to that of the college and univer-
sity. The development of the latter, all librarians know,
has not kept pace with that of the former. It has been
too much the tendency of such libraries to serve only their
own immediate communities the student bodies and the
faculties. This is due, no doubt in a large mesure, to the
entirely inadequate support that it receives. It would seem
that university authorities have sometimes failed to recog-
nize the real function of the university library, the larger
scope and function of its activities, the multiplicity of detail
connected with its administration. They have failed to
see that the library is really an institution in itself, with
great possibilities for state- wide service and that, as such,
it should receive liberal financial support; and we have the
conviction that the fault lies partly with the librarians in
not making the influence of their libraries felt more in the
state at large in not putting them upon a plane which
will command the respect which justly belongs to them.

ne wva rietyofc ret3716ric h, leaf

A New Variety of Cretaceous Decapod

from Texas

Abbyy 8 accuracy — page 6

Eugene S. Richardson, Jr.
Curator of Fossil Invertebrates

Fragments of the large crustacean, Enoploclytia walkeri (Whit-
field) are not infrequently reported from Texas. Localities are
listed by Stenzel (1944, p. 421) and by Rathbun (1935, p. 23, as
Palaeastacus walkeri), in the Fort Worth and Weno limestones of
the upper Albian, near the summit of the lower Cretaceous. The
geographic distribution as now known is within the outcrop belt of
these rocks and extends for about 250 miles from Denton County
on the north to Bexar County on the south.

The specimen discussed here was collected by Karl P. Schmidt
and Billy J. Anderson near the Bosque-McLennan County line, a
few miles north of China Springs, about in the middle of the known
distributional range of the species. The specimen had weathered
free but was found still in place in its impression on a massive
ledge outcropping on the gently rolling upland surface of the Washita
Upland (Grand Prairie). With it was associated an echinoid.

Order Decapoda
Family Erymidae

Genus Enoploclytia McCoy 1 849

Enoploclytia walkeri, var. schmidti, var. nov. Figure 108.

Type specimen. Chicago Natural History Museum no. P29710.

Description. The specimen collected by Schmidt and Anderson
consists of the fingers, palm (manus), wrist (carpus), and part of
the arm (merus) of an individual of the same size as the holotype
and slightly larger than the Weno specimen figured by Stenzel

(1944, pQ. 38); all are right chelipeds. The fixed finger is almost
entire and is somewhat longer in relation to the length of the
cheliped than in the Weno specimen referred to above; Whitfield's

Genus Enoploclytia McCoy 1849

Enoploclytia walkeri, var. schmidti, var. nov. Figure 108.

Type specimen. Chicago Natural History Museum no. P29710.

(1944, pQ. 38); all are right chelipeds. The fixed finger is almost
entire and is somewhat longer in relation to the length of the
cheliped than in the Weno specimen referred to above; Whitfield's

Abbyy 8 accuracy — page 7

445

tie wva rietyofc ret3716ric h, leaf 0012

REFERENCES

Adkins,W. S.

1923. Geology and mineral resources of McLennan County. Univ. Texas
Bull., no. 2340, 202 pp., map, 4 p|>.

1923. Geology and mineral resources of McLennan County. Univ. Texas
Bull., no. 2340, 202 pp., map, 4 pQs.

Rathbun, Mary J.

1935. Fossil Crustacea of the Atlantic and Gulf coastal plain. Geol. Soc.
Amer., Sp. Pap., no. 2, vii + 160 pp., 2 figs., 26 pys.

1935. Fossil Crustacea of the Atlantic and Gulf coastal plain. Geol. Soc.
Amer., Sp. Pap., no. 2, vii + 160 pp., 2 figs., 26 pgs.

Stenzel, H. B.

1944. Decapod crustaceans from the Cretaceous of Texas. Univ. Texas Bull.,
no. 4401, pp. 402-476, figs. 1-16, p[]s. 34-45.

1944. Decapod crustaceans from the Cretaceous of Texas. Univ. Texas Bull.,
no. 4401, pp. 402-476, figs. 1-16, pjs. 34-45.

Whitfield, Robert P.

1883. Paramijhrax? walked Whitfield, pp. 37, 38, p(]. 16, fig. ]a, pj. 17, fig.

|a, in White, C. A., Report, pp. 1-171 in Hayden, F. V., 12th Ann. Rept.
U. S. Geol. and Geogr. Surv. Terr, for 1878.

1883. Paramijirax? walkeri Whitfield, pp. 37, 38, p[ 16, fig. ]a, p[ 17, fig.

[]a, in White, C. A., Report, pp. 1-171 in Hayden, F. V., 12th Ann. Rept.
U. S. Geol. and Geogr. Surv. Terr, for 1878.

448

jewishprublemhow

^^^^1 ■ ■ ■ ^" 4^4^4^^*

DObrannchp leaf 0006

r ^^^^^^ ■ ^^ ■ ■■ B^rf ■ ■■ B^^^^ ■ ^^^^^^^^

T[jE JEWISH PROBLEM:

T||1e JEWISH PROBLEM:

Jew and that of Jews collectively. Obviously, no individual
should 0e subjected anywhere, by reason of the fact that he
is a Jew, to a denial of any common right or opportunity en-
joyed by non-Jews. But Jews collectively should likewise en-
j oy the same ri^ht and opportunity to live and develop as do

Jew and that of Jews collectively. Obviously, no individual
should 0e subjected anywhere, by reason of the fact that he
is a Jew, to a denial of any common right or opportunity en-
joyed by non-Jews. But Jews collectively should likewise en-
joy the same riSflht an d opportunity to live and develop as do
other B^ffifflps of people. This right of development on th0

other fgJjEJps of people. This right of development on thQ

Abbyy 8 accuracy — page 8

part of the group is essential to the full enjoyment of rights

jjevelopme

part of the group is eslential to the full enjoyment of rights

by the individual. For the individual is dependent for

|nt (and his happiness) in large part upon the de-
velop fnTCTI of the group of whch he forms a part. We can

scarcely conceive of an individual German or Frenchman liv-
ing and developing without some relation to the contempor-
ary German or French life and culture. And since death i|
not a solution of the problem of life, the solution of the Jew-
ish Problem necessarily involves the continued existence of
the Jews as Jews.

Councils of Rabbis and others have undertaken at times
to prescribe by definition that only those shall be deemed Jew0
who professedly adhere to the orthodox or reformed faith.
But in the connection in which we are considering the term,
it is not in the power of any single body of Jews or indeed of
all Jews collectively to establish the effective definition. The
meaning of the word Jewish in the term Jewish Problem
must 0e accepted a@ co-extensive with the disabilities which
it is our problem to remove. It is the non-Jews who create
the disabilities and in so doing give definition to the term Jew.
Those disabilities extend substantially to all of Jewish blood.
The disabilities do not end with a renunciation of faith,
however sincere. They do not end with the elimination,
however complete, of external Jewish mannerisms. The dis-
abilities do not end ordinarily until the Jewish blood has
been so thoroughly diluted by repeated intermarriages as to
result in practically obliterating the Jew.

And we Jews, by our own acts, give a like definition to
the term Jew. When men and women of Jewish blood suffer
because of that fact and even if they suffer from
quite different causes our sympathy and our help goes out
to them instinctively QJ whatever country they may live and
without inquiring into the shades of their belief or unbelief.
When those of Jewish blood exhibit moral or intellectual su-
periority, genius or special talent, we feel pride in them,

ilovel()]:>moi

by the individual. For the individual is dependent for

t (and his happiness) in large part upon the de-

velop^JSl °f ^ g rou P of whch he forms a part. We can
scarcely conceive of an individual German or Frenchman liv-
ing and developing without some relation to the contempor-
ary German or French life and culture. And since death i0
not a solution of the problem of life, the solution of the Jew-
ish Problem necessarily involves the continued existence of
the Jews as Jews.

Councils of Rabbis and others have undertaken at times
to prescribe by definition that only those shall be deemed Jew@
who professedly adhere to the orthodox or reformed faith.
But in the connection in which we are considering the term,
it is not in the power of any single body of Jews or indeed of
all Jews collectively to establish the effective definition. The
meaning of the word Jewish in the term Jewish Problem
must @e accepted a0 co-extensive with the disabilities which

it is our problem to remove. It is the non-Jews who create
the disabilities and in so doing give definition to the term Jew.
Those disabilities extend substantially to all of Jewish blood.
The disabilities do not end with a renunciation of faith,
however sincere. They do not end with the elimination,
however complete, of external Jewish mannerisms. The dis-
abilities do not end ordinarily until the Jewish blood has
been so thoroughly diluted by repeated intermarriages as to
result in practically obliterating the Jew.

And we Jews, by our own acts, give a like definition to
the term Jew. When men and women of Jewish blood suffer
because of that fact and even if they suffer from
quite different causes our sympathy and our help goes out
to them instinctively whatever country they may live and
without inquiring into the shades of their belief or unbelief.
When those of Jewish blood exhibit moral or intellectual su-
periority, genius or special talent, we feel pride in them,

Abbyy 8 accuracy— page 9

even if they have abjured the faith like Spinoza, Marx, Di0-

[2]

even if they have abjured the faith like Spinoza, Marx, Di§-

[2]

jewishpiDblemhowOObianrich, leaf

iii

HOW TO SOLVE IT

that by such domination only, does civilization advance.
Strong Rationalities assume their own superiority, and come
to believe that they possess the divine right to subject other

peoples to their sway. Soon the belief in the existence of
such a right becomes converted into a conviction that a duty
ex^ts to enforce it. Wars of aggrandizement follow as a
natural result of this belief.

that by such domination only, does civilization advance.
Strong Rationalities assume their own superiority, and come
to believe that they possess the divine right to subject other

peoples to their sway. Soon the belief in the existence of
such a right becomes converted into a conviction that a duty
ex^E ts t0 enforce it. Wars of aggrandizement follow as a
natural result of this belief.

■■

This attitude of certain nationalities is the exact corre-
lative of the position which was generally assumed by the
strong in respect to other individuals before democracy be-
came a common possession. The struggles of the eighteenth
and ni netee nth centuries both in peace and in war were de-
voted fETH elv to overcoming that position as to individuals.
In establishing the equal right of every person to develop-
ment, it became clear that equal opportunity for all involves
this necessary limitation: Each man may develop himself
so far, but only so far, as his doing so will not interfere with

the exercise of a like right by all others. Thus liberty came
to mean the right to enjoy life, to acquire property, to pursue

happiness in such manner and to such extent as the exercise
of the right in each is consistent with the exercise of a like
right by every other of our fellow citizens. Liberty thus de-
fined underlies twentieth century democracy. Liberty thu§
defined exists in a large part of the western world. And even
where this equal right of each individual has not yet been
accepted as a political right, its ethical claim is gaining rec-
ognition. Democracy rejected the proposal of the superman

who should ri§e through sacrifice of the many. It insists
that the full development of each individual is not only a

This attitude of certain nationalities is the exact corre-
lative of the position which was generally assumed by the
strong in respect to other individuals before democracy be-
came a common possession. The struggles of the eighteenth
and ni netee nth centuries both in peace and in war were de-
voted ^^ely to overcoming that position as to individuals.
In establishing the equal right of every person to develop-
ment, it became clear that equal opportunity for all involves
this necessary limitation: Each man may develop himself
so far, but only so far, as his doing so will not interfere with

the exercise of a like right by all others. Thus liberty came
to mean the right to enjoy life, to acquire property, to pursue

accepted as a political right, its ethical claim is gaining rec-
ognition. Democracy rejected the proposal of the superman

who should ri0e through sacrifice of the many. It insists

that the full development of each individual is not only a

Abbyy 8 accuracy — page 10

ri^ht, but a duty to society: and that our best hope for civil-
ization Qies not in [^iformity, but in wide differentiation.

The movements of the last century have proved that
whole peoples have individuality no less marked than that
of the single person; that the individuality of a people is ir-
repressible, and that the misnamed internationalism which
seeks the obliteration of nationalities or peoples is unattain-
able. The new nationalism proclaims that each race or peo-
ple, like each individual, has a right and duty to develop,
and that only through such differentiated development will
high civilization be attained. Not until these principles or
nationalism, like those of democracy are generally accepted,
will liberty be fully attained, and minorities be secure in

ri[Jht, but a duty to society: and that our best hope for civil-
ization flies not in ^iformity, but in wide differentiation.

[5]

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Analysis of Abbyy OCR Accuracy"

See other formats