U.S. patent application number 10/723370 was filed with the patent office on 2004-06-03 for email text checker system and method.
Invention is credited to Gross, Anthony A., Gross, John N..
Application Number | 20040107089 10/723370 |
Document ID | / |
Family ID | 32391926 |
Filed Date | 2004-06-03 |
United States Patent
Application |
20040107089 |
Kind Code |
A1 |
Gross, John N. ; et
al. |
June 3, 2004 |
Email text checker system and method
Abstract
A word processing tool is disclosed for checking the substance
and not merely the spelling of words provided by a user. The word
checker is capable of identifying potentially inappropriate word
choices so that unintentional errors are not introduced into
electronic text documents. The word checker can be implemented as a
stand-alone procedure, or integrated into a conventional
spell-checking program.
Inventors: |
Gross, John N.; (Woodside,
CA) ; Gross, Anthony A.; (San Jose, CA) |
Correspondence
Address: |
J. NICHOLAS GROSS, ATTORNEY AT LAW
726 DUBOCE AVE.
SAN FRANCISCO
CA
94117
US
|
Family ID: |
32391926 |
Appl. No.: |
10/723370 |
Filed: |
November 26, 2003 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10723370 |
Nov 26, 2003 |
|
|
|
09014414 |
Jan 27, 1998 |
|
|
|
Current U.S.
Class: |
704/10 |
Current CPC
Class: |
H04L 51/12 20130101;
H04L 51/14 20130101; G06Q 10/107 20130101; G06F 40/232 20200101;
G06F 40/253 20200101 |
Class at
Publication: |
704/010 |
International
Class: |
G06F 017/21 |
Claims
What is claimed is:
1. A method of checking the meaning of a word in an electronic
document comprising the steps of: retrieving said word to be
checked from said document; determining whether said word has been
inadvertently included in said document and has a particular
meaning that has been designated as potentially inappropriate for
use in said document.
2. The method of claim 1, further including a step of designating
words that have a meaning that is potentially inappropriate by
changing a value of a status field for such words in an electronic
dictionary.
3. The method of claim 2, wherein the status field for each of the
words can have one of N values, where N>2.
4. The method of claim 2, wherein those words that have a
potentially inappropriate meaning can be identified without input
from a user of such electronic dictionary.
5. The method of claim 2, wherein an identification of potentially
inappropriate words can take place at or about the time the
electronic dictionary is first generated.
6. The method of claim 1, further including a step of generating an
alert indicating that such word is potentially inappropriate.
7. The method of claim 6, further including a step of permitting a
user to change the value of the status flag for such word.
8. The method of claim 1, further including a step of verifying the
spelling of such word.
9. The method of claim 7, further including a step of generating a
list of substitute words in the event such word is mis-spelled, and
if a one of the substitute words is selected in lieu of such word,
determining whether the substituted word has a meaning that is
potentially inappropriate for use in a text document.
10. The method of claim 1, wherein a list of documents is checked,
and a list of potentially inappropriate words in such documents is
generated.
11. The method of claim 1, wherein said word is checked
substantially immediate in time after it is input into said
document by a user.
12. The method of claim 1, wherein said word has a first meaning in
a first context, and a second meaning in a second context, and
further including a step of determining whether said word has a
particular meaning that has been designated as potentially
inappropriate for use in either of the first or second
contexts.
13. The method of claim 2, further including a step of specifying a
threshold value which the status field must exceed in order for
said word to be identified as potentially inappropriate.
14. A method of permitting a user to simultaneously check both the
spelling and meaning of words in an electronic document comprising
the steps of: [a] retrieving a word to be spell checked and word
checked from said document; and [b] determining whether said word
has been spelled correctly; and [c] when said word has been spelled
incorrectly, presenting a first list of alternative words to said
user as replacements for said word, and permitting said user to
select a first replacement word for said word; and [d] determining
whether the meaning of said word or replacement word has been
designated as potentially inappropriate for use in a text document;
and [e] when said word has been designated as potentially
inappropriate, presenting a second list of alternative words to
said user as replacements for said word, and permitting said user
to select a second replacement word for said word.
15. The method of claim 14, further including a step (f):
permitting said user to select said word, the first replacement
word, or the second replacement word.
16. The method of claim 14, further including a step (f): repeating
steps [d] and [e] as necessary so that a word is selected that is
not designated as potentially inappropriate, or said user is
permitted to select a word by manual override.
17. The method of claim 14, further including a step of identifying
those words that have a meaning that is potentially inappropriate
by changing a value of a status field for such words in an
electronic dictionary.
18. The method of claim 17, wherein those words that have a
potentially inappropriate meaning can be identified without input
from a user of such electronic dictionary.
19. The method of claim 17, wherein an identification of
potentially inappropriate words can take place at or about the time
the electronic dictionary is first generated.
20. The method of claim 14, further including a step of generating
an alert indicating that such word is potentially
inappropriate.
21. The method of claim 14, further including a step of permitting
a user to change the value of the status flag for such word.
22. The method of claim 14, wherein said word is checked
substantially immediate in time after it is input into said
document by a user.
23. The method of claim 14, wherein said word has a first meaning
in a first context, and a second meaning in a second context, and
further including a step of determining whether said word has a
particular meaning that has been designated as potentially
inappropriate for use in either of the first or second
contexts.
24. A method of word checking an electronic document generated by a
word processing program under control of a user of said program,
said method comprising the steps of: storing word-checking status
information for said document, including a table containing
potentially inappropriate words used in said document; monitoring
interaction between said user and said word processing program to
identify idle editing periods; locating potentially inappropriate
words in said document during the idle editing periods, and
updating the word checking status information in response thereto
as needed.
25. The method of claim 24, wherein the table includes information
pertaining to the location in said document of any words that are
determined to be potentially inappropriate.
26. The method of claim 24, wherein a check of the spelling of such
word is also performed.
27. The method of claim 24, wherein said word has a first meaning
in a first context, and a second meaning in a second context, and
further including a step of determining whether said word has a
particular meaning that has been designated as potentially
inappropriate for use in either of the first or second
contexts.
28. The method of claim 24, further including a step of permitting
a user to specify a threshold which must be exceeded in order for
said word to be identified as potentially inappropriate.
29. A method of automatically word checking an electronic document
as it is generated by a user of a word processing program,
comprising the steps of: monitoring data input by said user during
a data input period that is intended to be included as a word in
said document; determining whether the word is potentially
inappropriate for use in said document; when the word is determined
to be potentially inappropriate, alerting said user; wherein a
result of said word checking is communicated to said user with
substantially minimal delay after the data input period for the
word.
30. The method of claim 29, wherein said result is communicated
before said user has completed data input during a subsequent data
input period intended for a subsequent word to be included in said
document.
31. The method of claim 29, wherein when the word is determined to
be potentially inappropriate said user is precluded from inputting
additional words in said document until corrective action has been
taken for such word.
32. The method of claim 29, wherein a check of the spelling of such
word is also performed, and wherein the result of this spelling
check is communicated to said user with substantially minimal delay
after the data input period for the word.
33. The method of claim 32, further including a step of generating
a list of substitute words in the event such word is mis-spelled,
and if a one of the substitute words is selected in lieu of such
word, determining whether the substituted word has a meaning that
is potentially inappropriate for use in a text document.
34. The method of claim 29, wherein said word has a first meaning
in a first context, and a second meaning in a second context, and
further including a step of determining whether said word has a
particular meaning that has been designated as potentially
inappropriate for use in either of the first or second
contexts.
35. The method of claim 29, further including a step of permitting
a user to specify a threshold which must be exceeded in order for
said word to be identified as potentially inappropriate.
36. A method of generating an electronic version of a dictionary
comprising the steps of: inputting data in electronic form to
create a set of words in computer readable form for said
dictionary; and providing a status field associated for each of
such words, which status field indicates whether such word has a
meaning that is potentially inappropriate for use in an electronic
document.
37. The method of claim 36, wherein the status field is modifiable,
and further including a step of modifying the status field in
accordance with the meaning of its associated word.
38. The method of claim 36, wherein the status field can be
modified to have any one of N possible values, where N>=2, and
where N corresponds to a degree of potential inappropriateness for
a word associated with such status field.
39. The method of claim 36, wherein each status field is associated
with the meaning of such word in a first context, and including a
further step of providing at least one additional status field for
indicating whether such word has a meaning that is potentially
inappropriate for use in a second context.
40. An electronic system for permitting a user to check the meaning
of words in an electronic document, said system including: a system
memory for storing a word checking software module; a processing
device for executing the word checking software module; wherein the
word checking software module is configured such that it can
perform the following operations: (i) retrieving said words to be
checked from said document; (ii) determining whether said word has
been inadvertently included in said document and has a particular
meaning that has been designated as potentially inappropriate for
use in said document.
41. The system of claim 40, further including an electronic
dictionary for storing said words, and wherein the word checking
software module designates words that have a meaning that is
potentially inappropriate by changing a value of a status field for
such words.
42. The system of claim 41, wherein the status field for each of
the words can have one of N values, where N>2.
43. The system of claim 41, wherein those words that have a
potentially inappropriate meaning can be identified without input
from said user.
44. The system of claim 41, wherein an identification of
potentially inappropriate words can take place at or about the time
the electronic dictionary is first generated.
45. The system of claim 40, wherein the word checking software
module is configured such that it can perform the additional
operation of generating an alert indicating that such word is
potentially inappropriate.
46. The system of claim 45, wherein the word checking software
module is configured such that it can perform the additional
operation of permitting said user to change the value of the status
flag for such word.
47. The system of claim 40, wherein the word checking software
module is configured such that it can perform the additional
operation of verifying the spelling of said words.
48. The system of claim 47, wherein the word checking software
module is configured such that it can perform the additional
operation of generating a list of substitute words in the event
such word is mis-spelled, and if a one of the substitute words is
selected in lieu of such word, determining whether the substituted
word has a meaning that is potentially inappropriate for use in a
text document.
49. The system of claim 40, wherein the word checking software
module is configured such that it can perform the additional
operation of checking a set of documents, and generating a list of
potentially inappropriate words in such documents.
50. The system of claim 40, wherein said word is checked
substantially immediate in time after it is input into said
document by a user.
51. The system of claim 40, wherein said word has a first meaning
in a first context, and a second meaning in a second context, and
wherein the word checking software module is configured such that
it can perform the additional operation of determining whether said
word has a particular meaning that has been designated as
potentially inappropriate for use in either of the first or second
contexts.
52. The system of claim 40, wherein the word checking software
module is configured such that it can perform the additional
operation of specifying a threshold value which the status field
must exceed in order for said word to be identified as potentially
inappropriate.
53. An article of manufacture comprising: an electronic dictionary
including a set of words and an associated status field for each of
the words for indicating a level of potential inappropriateness for
such word when used in an electronic text document; a word checking
routine executable by a computing device and for checking whether
words in the electronic text document are inadvertently included,
and which routine is configured to perform the following
operations: [a] retrieving a word to be word checked from an
electronic text document; and [b] accessing the electronic
dictionary to determine whether the level of potential
inappropriateness of such word exceeds a threshold value specified
for such document; and wherein the electronic dictionary and word
checking routine are embodied in a computer readable media.
54. The article of claim 53, further including a spell checking
routine for determining whether such word has been spelled
correctly.
55. The article of claim 54, wherein the spell checking routine
also presents a first list of alternative words as replacements for
the word when it is mis-spelled, and permits a first replacement
word to be substituted for the word.
56. The article of claim 55, wherein the word checking routine is
also configured to determine whether the meaning of the replacement
word exceeds the threshold value for such document.
57. The article of claim 53, wherein the word checking routine is
also configured to permit a user of the routine to change the value
of a status field for any words in the electronic dictionary.
58. The article of claim 53, wherein the word checking routine is
also configured to check words substantially immediate in time
after they are input into the document by a user.
59. The article of claim 53, wherein the electronic dictionary has
a first status field for indicating a level of appropriateness for
a word in a first context, and a second status field for indicating
a level of appropriateness for such word in a second context, and
the word checking routine is configured to check whether any words
have a particular meaning that has been designated as potentially
inappropriate for use in either of the first or second
contexts.
60. The article of claim 53, wherein the computer readable media
includes any of the following: a floppy disk, a hard disk, a
CD-ROM, a magnetic tape, or a non-volatile semiconductor
memory.
61. A system for permitting a user to check the meaning of words in
an electronic document, said system including: a computer readable
storage structure for storing a word checking software module, such
module being incorporated as part of a word processing program that
can be executed by a computer processing device; wherein the word
checking software module is configured such that when it is
executed by the computer processing device it can perform the
following operations: (i) retrieving said words to be checked from
said document; (ii) determining whether said word is inadvertently
included and has a particular meaning that has been designated as
potentially inappropriate for use in said document.
62. The system of claim 61, further including an electronic
dictionary for storing said words, and wherein the word checking
software module designates words that have a meaning that is
potentially inappropriate by changing a value of a status field for
such words.
63. The system of claim 62, wherein the status field for each of
the words can have one of N values, where N>2.
64. The system of claim 61, wherein the word checking software
module is configured such that it can perform the additional
operation of generating an alert indicating that such word is
potentially inappropriate.
65. The system of claim 61, wherein the word checking software
module is configured such that it can perform the additional
operation of permitting said user to change the value of the status
flag for such word.
66. The system of claim 61, wherein the word checking software
module is configured such that it can perform the additional
operation of verifying the spelling of said words.
67. The system of claim 66, wherein the word checking software
module is configured such that it can perform the additional
operation of generating a list of substitute words in the event
such word is misspelled, and if a one of the substitute words is
selected in lieu of such word, determining whether the substituted
word has a meaning that is potentially inappropriate for use in a
text document.
68. The system of claim 61, wherein the word checking software
module is configured such that it can perform the additional
operation of checking a set of documents, and generating a list of
potentially inappropriate words in such documents.
69. The system of claim 61, wherein said word is checked
substantially immediate in time after it is input into said
document by a user.
70. The system of claim 61, wherein said word has a first meaning
in a first context, and a second meaning in a second context, and
wherein the word checking software module is configured such that
it can perform the additional operation of determining whether said
word has a particular meaning that has been designated as
potentially inappropriate for use in either of the first or second
contexts.
71. The system of claim 61, wherein the word checking software
module is configured such that it can perform the additional
operation of specifying a threshold value which the status field
must exceed in order for said word to be identified as potentially
inappropriate.
72. An electronic dictionary embodied in computer readable form for
use in a word processing program, said dictionary including: a set
of words represented in computer readable form; a status field
associated for each of such words, which status field facilitates a
determination of whether such word, when inadvertently included in
an electronic document, has a meaning that is potentially
inappropriate for use in such document.
73. The dictionary of claim 72, wherein the status field is
modifiable.
74. The dictionary of claim 73, wherein the status field has one of
N possible values, where N>2, and where N corresponds to a
degree of potential inappropriateness for a word associated with
such status field.
75. The dictionary of claim 73, wherein each status field is
associated with the meaning of such word in a first context, and
including at least one additional status field for indicating
whether such word has a meaning that is potentially inappropriate
for use in a second context.
76. A method of checking the meaning of a word in an electronic
document comprising the steps of: retrieving said word to be
checked from said document; determining whether said word has a
particular meaning that is likely to be offensive to persons
intended to read such text document.
77. The method of claim 76, further including a step of designating
words that have a meaning that is likely to be offensive by
changing a value of a status field for such words in an electronic
dictionary.
78. The method of claim 77, wherein the status field for each of
the words can have one of N values, where N>2.
79. The method of claim 76, further including a step of generating
an alert indicating that such word is likely to be offensive.
80. The method of claim 76, further including a step of verifying
the spelling of such word.
81. The method of claim 80, further including a step of generating
a list of substitute words in the event such word is mis-spelled,
and if a one of the substitute words is selected in lieu of such
word, determining whether the substituted word has a meaning that
is likely to be offensive.
82. The method of claim 76, wherein said word is checked
substantially immediate in time after it is input into said
document by a user.
Description
FIELD OF THE INVENTION
[0001] The present invention relates to authoring tools that can be
used in connection with contemporary word processing programs. In
particular, the present invention provides an author of an
electronically drafted document with a word checker that checks for
and identifies inappropriate word choices in such document based on
a sensitivity scheme of the user's choosing so that such words may
be modified if necessary.
BACKGROUND OF THE INVENTION
[0002] The prior art is replete with word processing programs,
including a couple of contemporary favorites, Microsoft WORD and
Novell's Wordperfect, that are used by a substantial portion of
IBM-compatible computer users. These programs are used in known
ways for permitting authors to create electronic text (and
graphics) documents. As a part of such word processing program, a
spell-checking routine is almost always included to help authors
reduce the number of unintentional text errors in such documents. A
number of prior art patents are directed to this feature, and a
reasonable background of the same is described in U.S. Pat. Nos.
5,604,897 to Travis and U.S. Pat. No. 5,649,222 to Mogilevsky, both
of which are hereby incorporated by reference.
[0003] It is apparent, however, that spell-checking routines
associated with such word processing programs have a number of
limitations. Key among these is the fact that they cannot determine
whether a particular word choice, while accurately spelled, is
nevertheless perhaps inappropriate for the particular context
within a particular document. As an example, many words that may be
intended by a drafter (such as the words "ask," "suit," "public,"
etc.) can be transformed into potentially offensive words merely by
changing a single letter in such words, transposing a few letters,
or by mistakenly adding or dropping a letter. These transformed
words, however, will still pass the spell-checking facility,
because many of them include even a number of offensive words as
part of their standard dictionary. For example, the word "ask" may
be inadvertently written as "ass," and unless the message is
intended to discuss issues pertaining to certain members of the
animal kingdom, it is likely to be an inappropriate word choice. If
these inadvertent mistakes are not caught by the drafter during a
later review, they will be included in such document and
potentially communicated to one or more third parties. Depending on
the severity of the mistake, the receiving audience, and the scope
of the distribution of the document, the consequences may range
from minor embarassment to substantial financial loss from lost
future business with such third party.
[0004] The possibility of such errors is increasing each day
because of a number of driving factors, including the fact that
standard dictionaries for word processors are growing in size to
accommodate the largest number of words of course in a particular
language. While one solution may be to not include such words in an
electronic dictionary in the first place, this result makes the
creation of such dictionaries more complicated because an initial
censoring must be done before the words are even translated into
electronic form. Moreover, this solution does not help the user to
identify inappropriate words that may be skipped over during a
spell-checking routine.
[0005] Another factor leading to increase in electronic word choice
errors is the fact that many electronic documents are never reduced
to a physical form before being disseminated. In many instances a
glaring error is caught by a human inspection of a printed page
before it is sent out. The so-called "paperless office" while
improving efficiency and reducing waste also naturally causes a
larger number of inadvertent message errors in text documents.
Additional errors can even be induced by spell-checkers because
when they detect a mis-spelled word, they will often provide a menu
of potential word choices as replacements, and it is remarkably
easy to select an inappropriate word choice from such menu, again
merely by accident. Such errors of course will not be detected
because the document is erroneously considered to be "safe" by many
users after spell-checking has completed and they will not check it
again. In other words, some facility for checking the spell-checker
dynamically is also desirable, but does not exist at this time.
[0006] There is some facility in the prior art for permitting users
to create so-called "exclusion" dictionaries for analyzing text
documents. An example of such kind of system is illustrated in U.S.
Pat. No. 5,437,036 to Stamps et. al. which is incorporated by
reference herein. A drawback of this approach, however, lies in the
fact that it requires the user to both divine and manually input
all the potential mis-spellings that could occur, and even if they
had the time, there are obviously an endless variety that might
never be considered by such user. For example, a user may not have
the foresight to notice that a simple transposing of two characters
(a common error) may generate a word that is extremely offensive.
Furthermore Stamps et. al. do not appear to contemplate the
possibility that the act of rendering a document "spelling" error
free may itself generate unintended word selection errors. As such,
therefore, Stamps et. al. is not truly a "word" checker, but,
rather, an enhanced spell checker that has been sensitized to a
particular user's poor spelling habits. While it incidentally
determines whether a word is perhaps not the intended choice of the
author (i.e., that the word does not have a particular meaning), it
does not perform the important step of determining the precise
meaning of the word, and in particular whether the word also has a
potentially inappropriate meaning as well.
[0007] A few methods for proof-reading electronic documents are
also known in the art. A U.S. Pat. No. 4,674,065 to Lange et. al.,
also incorporated by reference herein, describes a technique for
detecting word context errors in a document. This technique seems
limited to homophones however (for example, it knows to see if a
user intended to use the word "course" instead of "coarse") and is
not generally applicable to the problem of determining
inappropriate use of language in documents. For example, unless a
particularly offensive word has a homonym, Lange et. al. would not
even detect such word as being a problem. The approach of Lange et.
al. further requires a fair amount of computational complexity,
since it must analyze the text preceeding and following after a
word and use a complicates set of syntax rules to determine whether
the word is being used in context correctly. This fact alone makes
it essentially unusable for most contemporary word processing
programs which utilize background spell checking, dynamic
spell-checking, etc.
[0008] Finally, a U.S. Pat. No. 4,456,973 to Cargren et al., and
also incorporated by reference herein, discusses the use of an
electronic word dictionary that has an associated code field for
indicating the level of comprehensibility of such word. For
example, the word "abandon" is coded with a numerical designation
6, indicating that the word is probably understandable by children
at the 6.sup.th grade level. Cargren et al., however, do not appear
to address the more general problem of identifying text that has
been inadvertently mis-spelled by an author, and which is likely to
be inappropriate. In other words, the Cargren al. approach presumes
that the user has correctly input the word in question, and unless
the word is coded with a rating below that of the intended grade
group of children, it is not flagged in anyway. It is apparent that
this method of encoding is fairly impractical for use in an
electronic dictionary intended to be used by an adult population,
because adults are not classified in this way. In fact, if a target
audience of a document is intended to be primarily adults, then the
Carlgren et al. approach would not flag any words at all, because
they would probably be presumed to be operating at the highest
level of education (12), thus rendering this type of filtering
essentially useless. In addition, there is no facility mentioned by
Cargren et al. for detecting words that are likely to be offensive,
even if consciously selected by the author. For example, the use of
the word "dame" may be consciously selected but nevertheless
undesirable in communications in which the intended audience is
primarily adult women. A drafter of an electronic document may
desire to be notified of such potentially offensive words if they
are known to be sensitive.
SUMMARY OF THE INVENTION
[0009] An object of the present invention therefore is to reduce
the number of unintentional inappropriate word choices within
electronic documents that would otherwise go unnoticed using
conventional word processing document checking tools so as to
improve the integrity and accuracy of such documents;
[0010] Another object of the present invention is to provide a
system and method that is easily and seamlessly integratable into
conventional word processing document checking tools so as to
enhance the performance of such tools;
[0011] A related object of the present invention is to provide a
system and method for filtering and verifying the contents of one
or more electronic documents to determine the presence of
potentially inappropriate and unintended word choices;
[0012] Yet a further object of the present invention is to reduce
the number of intentional but unknowingly inappropriate word
choices within electronic documents that would otherwise go
unnoticed using conventional word processing document checking
tools;
[0013] A related object is to improve the performance of present
day word processing document checking tools by providing an
additional verification tool that confirms the appropriateness of
the selections made by such prior art checking tools;
[0014] Another object of the present invention is to permit a user
of a word processing program to selectively control the level of
sensitivity to be used for determining whether words in an
electronic document are potentially inappropriate;
[0015] Still another object of the present invention is to permit
an author of an electronic document to have the words of such
document analyzed and processed by a number of context filters of
the author's choosing to reduce the number of potential
inappropriate words in such document.
[0016] These and other objects are achieved by the present
invention which includes a word checking software routine
(implementable as a stand-alone program or integrated with a
conventional spell checker) that checks the meaning of words in an
electronic document authored by a user. When word-checking for the
document is desired by the user, the words from the document are
retrieved one at time, and checked against entries in an electronic
dictionary to determine whether they have a particular meaning that
has been designated as potentially inappropriate for use in a text
document. The determination is made based on comparing a threshold
sensitivity level (which can be controlled by the user) with an
appropriateness rating found in one or more status fields
associated with the word. If the word in question has a rating
higher than the threshold set by the user, an alert is provided to
indicate such result.
[0017] In another embodiment, multiple status fields are used. The
status fields can be used essentially as multiple context filters
for assisting an author in reducing the number of potentially
inappropriate words as they may be conveyed to multiple intended
audiences. The ratings for the words in any of the status fields
can have any range of values and are preferably coded at the time
the words are placed into the electronic dictionary. They can also
be modified by the user, later, if desired.
[0018] The present invention is completely integratable with a
conventional spell-checking program, so that the spelling of a word
can also be analyzed before its meaning is also checked. When a
mis-spelled word is found, a user can select from a list of
accurately spelled substitute words instead, but such substitute
word is also checked to see if has a meaning that is potentially
inappropriate for use in a text document.
[0019] Another variation of the present invention permits a user to
specify a set of documents to word-check, and to generate an output
indicating the results of such check.
[0020] In yet another embodiment, the meaning of the words in the
document are checked during an idle state of the word processing
program, so that the operation of checking the entire document at
the user's request at a later time can be reduced in time because
it will already have been partially completed.
[0021] To further reduce errors, another embodiment of the present
invention checks the meanings of words substantially immediate in
time after they are input into the document by the user. This can
ensure that a potential word problem is in fact brought to the
user's attention automatically and without the need for a further
command from the user.
[0022] The electronic dictionary used in the present invention
includes a series of records corresponding to words and their
associated status field values. The data for these records are
input in electronic form to create a set of words in computer
readable form for the dictionary, along with status fields
associated for each of such words. Again, as mentioned above, the
status fields indicate whether such word has a meaning that is
potentially inappropriate for use in a particular context. As the
dictionary is constituted, any number of status fields can be used
to correspond to different sensitivity ratings for different target
audiences.
[0023] The present invention can be included within a word checking
software module that is itself embodied and transferred in computer
readable form, such as a floppy disk, a hard disk, a CD-ROM, a
magnetic tape, or a non-volatile semiconductor memory. In this way,
a new type of computer system is effectuated which permits an
author to perform a word checking operation on an electronic text
document generated by a conventional word processing program
running on such system.
BRIEF DESCRIPTION OF THE DRAWINGS
[0024] FIG. 1 is a simplified block diagram of a word processing
system embodying the teachings of the present invention;
[0025] FIG. 2 is a flow chart illustrating the operation of a
document word checking tool implemented in accordance with the
teachings of the present invention;
[0026] FIG. 3 is a flow chart illustrating a variation of the
present invention adapted for checking words in a number of
electronic documents specified by a user.
DETAILED DESCRIPTION OF THE INVENTION
[0027] FIG. 1 illustrates a physical implementation of a preferred
embodiment of a system employing the present invention. A user can
interact with a word processing program 10 loaded in a system
memory of a conventional computing system, such as a personal
computer. Such programs typically include a spell checking routine
or module 20, broken out visually in the present figure for
clarification. The inventive routine of the present invention is
also illustrated as word checker routine or module 30.
[0028] Both spell checking routine 20 and word checker routine 30
have access to and can store/retrieve entries from a standard
electronic dictionary 40, which, again, is well-known in the art.
It is understood, of course that spell checking routine 20 and word
checker routine 30 may be embedded as one software module of
program 10, or alternatively, may be constituted as separate
programs that interacts with program 10 and each other. Those
skilled in the art will appreciate that a number of separate
modules (i.e., for controlling printing, document loading, etc.) of
program 20 are not shown, but they are not material to the
teachings of the present invention. A document 40 is also shown,
and this item can be generated and controlled by a user using word
processing program 10 in a conventional manner well-known in the
art.
[0029] Again, while the present illustration depicts the various
routines, documents and dictionaries as separate entities, it is
understood that this is a simplified description intended to convey
the key concepts of the present invention. During normal operation
of word processing program 20 these various entities are typically
loaded from a non-volatile memory storage device (such as a
magnetic hard disk) and then coexist in a physical and logical
sense within the same system memory.
[0030] A flowchart of the operation of a preferred embodiment of
the present invention is depicted in FIG. 2. When word checker
routine 30 is invoked at step 210, it begins to retrieve and
analyse words from document 10 in a manner well-known in the art
from similar techniques used in conventional spell-checkers. In one
variation spell checker routine 20 and word checker routine 30 are
"integrated" in the sense that they are invoked together and
inter-operate with each other. In this variation, the word from the
document is first spell-checked in the manner known in the prior
art. For example, if the word is not found in electronic dictionary
50 at step 220, spell checker routine 20 is invoked at 225. Again,
using conventional techniques, a set of suitable alternative words
are presented to a user at step 230 as replacements for the
unidentified word.
[0031] At this point the method of the present invention is not
different in any material fashion from prior art spelling programs.
At step 235, however, a substantial difference is found in the fact
that the present invention does not blindly permit the user to
select one of the alternatives presented at step 230. As mentioned
above, there is a non-insignificant chance of error presented when
large lists of words are presented to users during spell-checking
routines. It is altogether too easy to select the wrong word, and
worse yet, a very inappropriate word. For this reason, at step 235,
the word selected by the user is also checked to see if it is
flagged as potentially context restricted. In this manner, errors
are further reduced, and an additional level of accuracy is
achieved.
[0032] In one implementation the information concerning a rating of
the restricted status for the word can be stored in an electronic
field associated with a record for such word in a dictionary
database, as shown in the breakout of the diagram for dictionary
50. Each entry 51 in dictionary 50 includes a word and one or more
associated status fields (SF.sub.1, SF.sub.2 . . . SF.sub.n),
which, in its simplest form, can be a single bit field that is
either set or not set, depending on whether the word has been
previously designated as potentially inappropriate. This status bit
can be set either during the creation of dictionary 40, or at the
request of a user later on, much in the same way new words can be
added to auxiliary or supplemental dictionaries used in
conventional spell-checking programs.
[0033] In another embodiment, status field SF.sub.1 can be
allocated sufficient coding bits so that it may have any one of N
possible values, where N represents a rating indicating a relative
ranking of inappropriateness. This feature allows a degree of
grading the words of a vocabulary in shades of grey so to speak,
and permits more precise control by the user of the level of
inappropriateness that can be tolerated within any particular
document. For example, using a scale from 1-10, words that are
known and commonly used in a vulgar or offensive fashion can be
given a 10 value rating, while words that are commonly understood
in only harmless contexts can be given a 1 rating. Other words that
are not as fixed in meaning or easily classifiable can be given
value ratings in between this range, again, depending on their
common usage within a particular target audience. Associated status
fields SF.sub.1, SF.sub.2, etc., can be used to generate different
and unique sets of inappropriate words for different intended
recipient groups. While only three status fields are shown to
simplify the present discussion, it is understood that any number
may be used in connection with an electronic dictionary. For
instance, the ratings found in field SF.sub.1 can be based on word
meanings for a typical U.S. based English speaking audience, while
SF.sub.2 might be used to indicate the rating for the same word for
a different target audience (i.e., a different language speaking
group, a different geographic group within the same country, or an
audience with specific sensitivities to certain types of words).
The ratings for status fields SF.sub.1, SF.sub.2, SF.sub.3 may be
derived from survey results, from polling the intended audience
populations, from analyzing other published literature materials
for such intended groups, from organizations specializing in
advertising and marketing, etc.
[0034] One advantage of the present invention over the prior art
lies in the fact that the ratings of dictionary words 41 can be
pre-programmed, instead of requiring a user to inspect each entry
and designate the status manually. It is expected that such ratings
could be generated by persons skilled in the art of contemporary
language usage, or persons having particular knowledge or
understanding of terminology used in specific fields, or by
specific target audience groups, using some of the reference
materials (e.g. surveys and the like) described immediately above.
The words and associated ratings can be put into computer readable
form at the time of the creation of the dictionary by the vendor of
word processing program 10, or by a supplier of dictionary 50 using
a conventional database structure with accessible records or the
like.
[0035] It is equally feasible, of course, that an entirely separate
electronic dictionary 50' may be designated instead for potentially
context restricted words. In this kind of approach step 235 merely
checks such separate dictionary 50' to see if the word has been
included in the class of potentially inappropriate words. In any
event, a number of implementations may be employed for designating
potentially offensive words and the present invention is not
limited to any of the examples shown herein.
[0036] In the event the present invention is not embodied in a
routine that runs concurrently with spell checker routine 20, it is
apparent that steps 220, 235 and 230 would not occur. In other
words, in a stand-alone implementation where word checker routine
30 is invoked as another typical word processing tool within
document 40, the present routine progresses directly from step 215
to step 235 where the word is checked. The key aspect of the
present invention lies in the fact that the meaning or substance of
words, not simply the spelling of such words, are determined and
verified. In a preferred embodiment, a user of word processing
program 10 is able to control both: (1) the range of status fields
SF.sub.1, SF.sub.2, etc. to be checked, as well as (2) a threshold
level value for such field that should be used in determining
whether a particular word should be flagged. These control
mechanisms for the user can take the form of menu choices
implemented in well-known fashion in typical contemporary
spell-checking programs. In this manner, a particular user document
40 can be subjected to a variety of language "filters" and with
varying levels of sensitivity to provide additional functionality
and benefits. For example, some U.S. companies doing business in
foreign countries nevertheless communicate and advertise in English
to the local population, because of the ever increasing English
fluency of such countries. Documents intended for such audiences
can be subjected to filtering not only for English words that are
deemed inappropriate by U.S. citizens, but also which have a
negative meaning when transliterated into the language of the
country in question. Other users may find it useful to designate
one status field simply as a filter for those words that they know
are prone to over-use or misuse to increase their writing clarity
and proficiency. Accordingly, a user of program 10 can decide to
utilize any one or more of such filters in connection with the word
checking of document 40, and the selection of such filters can be
accomplished in any one of several known ways (i.e., highlighting
desired filters presented in the form of a list, specifying them by
name, etc.) In addition, the user can also determine (if desired)
the threshold level of sensitivity to be used during the checking
of words in document 40. In other words, when a particular status
field SF.sub.1 can have a value ranging from 1 to 10, the user can
specifically indicate that only words having a status field value
in excess of 8 should be considered as potentially
inappropriate.
[0037] In one sense, the present invention behaves as a superior
spell-checker, because it catches even correctly spelled words that
are nonetheless incorrect (from the user's intent perspective)
because they are based on other words inadvertently mis-spelled
originally by the user. In other words, the user had a word with
spelling "xyzzy" in mind, but this word was inadvertently input as
"xyyzy." If"xyyzy" (the transformed embodiment of"xyzzy") is
nevertheless a word in a spell-checking dictionary, the prior art
techniques will never detect this mistake as a true error, even
though it is undesirable from the user's perspective, and perhaps
worse, may lead to significant embarrassment if undetected. By
permitting a user to designate sets of words that should be
screened (even if correctly spelled) the present invention affords
a significantly higher level of confidence to the user that
documents will be generated error and embarassment free. This kind
of tool is especially helpful because of the fact that electronic
communications occur in rapid fashion, and dissemination of
electronic documents can take place with little chance or
opportunity for correcting mistakes.
[0038] In any event, should a match occur for the word in question
indicating that it falls within the parameters of the user's
specification for what should be flagged as an inappropriate word,
an alert is given to the user at step 240. The alert can take the
form of highlighting the word in the text of the document with an
accompanying warning that can be either visual, audible or both.
The alert can also indicate the identity of the filter that was
triggered by the word when more than one status field is used
during the word checking process. Again, the precise implementation
of the warning is not critical, and any one of many known methods
in the art can be used.
[0039] At this point, the user can decide at step 245 whether the
word is truly inappropriate, and if so, the user can provide a
substitute word, or override the program to keep the word as is.
The substitute word can be provided directly by the user, or a set
of suitable alternatives can be presented in the same way
conventional spell checking program 20 provides users with such
lists. Again, this substitute selection is nevertheless also
inspected and verified for its appopriateness by virtue of the fact
that the routine branches back to step 220. In this manner, the
potential for erroneous inclusion of offensive or inappropriate
language is further minimized.
[0040] In the event the user decides to stick with the word, an
option can be presented to the user at step 250 for reducing the
value of the status field for the word so that it will not be
identified as a potentially inappropriate word during a subsequent
check (unless a lower threshold is used of course). This might be
desirable, for example, in certain fields of use where ostensibly
inappropriate words may be nevertheless otherwise acceptable for
use in common communications. At step 255, therefore, a user can
modify the status of the word in dictionary 50 or 50' as the case
may be.
[0041] Those skilled in the art will appreciate that the present
invention can be used in a number of environments where documents
are electronically drafted and have electronic text, including with
spreadsheet programs, database programs, e-mail programs, etc. Many
of these programs have spell-checking capabilities built in, and
the present invention can piggy-back directly on such
capability.
[0042] Moreover the present invention can be used for batch
inspection and verification of electronic documents, as shown in
FIG. 3. Instead of working directly with an open document 40, for
example, an electronic file to be word checked can instead be
specified at step 305 by a user. This type of capability is common
in word-processing programs 10 today, and permits a user to select
a number of files that are to to undergo a particular operation
without opening them. For example, from the FILE OPEN pulldown
option in WORDPERFECT, a print operation can be effectuated on a
user specified list of files. In this same way, the present
invention as embodied in a word checker routine 30 can operate on a
number of files at shown at step 310. As each word is parsed at
step 315, an optional spell checking routine 320 first determines
if there is mis-spelling of such word, and, if so, tags the word as
such at step 325. If the word is otherwise identifiable because it
is in dictionary 40 or alternate dictionary 40', it is context
checked at step 330 in the manner described above and with the same
degree of available control by the user. Each word identified as
potentially inappropriate is flagged and tagged at step 335 before
the routine returns to process the next word. At the end of the
file word checking operation, a list of mis-spelled words and
identified potentially misappropriate words is generated for the
user's review. This list can be displayed visually, generated as a
text file, etc. so the user can determine whether there are
potential problems in disseminating such electronic files, and,
take corrective action if necessary.
[0043] It will be apparent to those skilled in the art that the
present invention can be implemented using any one of many known
programming languages such as C, C++, Pascal, etc. The details of
the specific implementation of the present word checker will vary
depending on the programming language used to embody the above
principles. Furthermore, while the present discussion has revolved
around IBM compatible machines it is equally clear that the present
invention is not limited to such environments and can be used with
Apple compatible machines, or for that matter with any machine that
includes a word processor, including personal digital assistants
and similar portable computing devices.
[0044] Yet another variation of the present invention makes use of
a concept disclosed in the Mogilevsky reference discussed above. In
this manifestation, spell checking routine 20 and word checker
routine 30 (whether integrated or standalone) can be implemented in
a well-known fashion that improves performance by having them
operate while there is no interaction going on between the user and
word processing program 10. In other words, the routine can take
advantage of otherwise idle time by checking the document and other
open documents if the user is not actively engaged in an operation
requiring immediate attention by word processing program 20. It is
apparent that such variations of the present invention could be
beneficial in a number of environments.
[0045] Another useful variation of the present invention can be
implemented to take advantage of another well-known concept in the
art, and that is the use of "dynamic" text checking. For example,
the Travis reference above explains how words can be spell-checked
on the fly, i.e., as the user is typing in many conventional word
processing programs. This feature permits a user to enter a word,
and then have it checked while he or she is still typing, and while
their attention is still focused on the document. The spell
checking routine 20 automatically substitutes the closest choice it
finds for commonly misspelled words; for example, the type word
"teh" is changed rapidly to, "the" while the user is perhaps still
entering text. In a similar vein, a dynamic word-checking feature
can be implemented, and it will be apparent to those skilled in the
art that this approach is merely a variation of the procedure
described in FIG. 2, except that the words retrieved at step 215,
are fetched in a background task which is running even while the
user is engaged in normal text entry, editing, etc. From the user's
perspective, the word-checker routine 30 merely appears as an
add-on verification tool feature to spell-checker 20, with
additional functionality and benefits for reducing document errors.
To minimize errors, the result of the word-check is communicated as
quickly as possible to the user, subject to system performance
limitations of course. In any event, with typical computing devices
it is expected that such result can be communicated before the user
has finished completing data entry of another word that must be
checked as well. In some application environments it may be
desirable to disable further data input once it has been determined
that a potentially inappropriate word has just been entered.
[0046] One environment where the dynamic variation is especially
useful in e-mail systems where people rarely spell-check a text
message after it is composed. Accordingly, the ability to identify
and alert users automatically of potential errors is particularly
helpful in this area.
[0047] In a general sense the present invention can be seen as a
useful tool for not only reducing unintentional errors in
electronic text documents, but also some intentional errors that
are not necessarily understood or appreciated by the user of a word
processing program 20. In other words, a word input by a user may
be classified by in an number of ways including: (1) an intentional
selection that is appropriate; (2) an intentional selection that is
inappropriate; (3) an unintentional selection that is
inappropriate; or (4) an unintentional selection that is
appropriate (in the sense that it is perhaps non-offensive even if
it is inapposite for the context). Of these classifications, it is
clear that the first category (1) do not require any remedial
action. It is equally clear that the category (4) mistakes are
perhaps undesirable, but are not easily identifiable, and the
computational complexity involved in doing so may not be
worthwhile. For example, a sentence that reads "I gave my him
jacket" instead of "I gave him my jacket" is not accurate but is
not offensive.
[0048] The present invention, however, is a simple, fast and
cost-effective tool that can be used to easily identify the
category (2) and (3) errors above. For the category (3) items, it
is a rather simple matter as described above to identify and
classify entries in an electronic dictionary with varying status
field values reflecting their common usage in a particular context.
Any number of potentially offensive, crude, vulgar, obscene or
inappropriate words are included in an electronic dictionary for
the sake of completeness (and for ease of integration from text to
electronic form) but it is rarely the case that they are desired to
be used in communication and they can be identified by setting the
associated status field to a high value. The category (2) items are
somewhat more subtle, but it is apparent that some words, even if
consciously selected, may nevertheless be inappropriate because of
a number of cultural considerations, or simply because they are
generally misused by the general population of authors. As an
example, a considerable amount of attention has been focussed
recently in the U.S. on eliminating gender specific vocabulary
except where absolutely necessary. A business entity attempting to
sell products or services to a primarily female based clientele may
benefit from a word checking tool that is sensitive to potential
gender issues raised by language found in documents intended to be
communicated to such clientele to reduce the possibility of an
miscommunication. Many ethnic groups have particular words or
phrases that are considered inappropriate or offensive by them,
even if they otherwise understood as inoffensive by other groups.
For example, Chevrolet automobiles designated "NOVA" did not sell
well in Latin American countries, because the term as translated
into Spanish means "does not go." In addition, many foreigners
attempting to write in English often confuse homonyms and
substitute the wrong word choice (for example, weather and
whether). Similar examples from other fields will be apparent to
those skilled in the art. For any such areas, it would be an easy
task for a person skilled in the field of contemporary language and
familiar with such potentially offensive vocabulary in such
specific area to flag such entries in electronic dictionary 50 by
using one of the associated status fields (SF.sub.1, SF.sub.2,
SF.sub.n, etc.), and/or to assemble and create a separate
customized electronic dictionary of words that potentially
implicate inappropriate language for a particular audience. These
dictionaries could supplement traditional electronic dictionary 50
and be invoked as needed by a user of word processing program 10,
so that any one or more of a number of word checking verifications
or document filtrations could be performed depending on the target
audience intended to recieve such document. Such additional
dictionaries, or additional status field values for a particular
filter, could be created in a number of ways, including as
additional incorporations in the original dictionary 50, or by the
vendor or user of computer program 10 modifying dictionary 50 at a
later time.
[0049] Finally, word processing program 10, spell-checking program
20 and word-checking program 30 described above can be embodied in
well-known ways in an article of manufacture such as in traditional
computer-readable media commonly used to transport such programs.
These can include a floppy disk, tape, hard disk, CD-ROM or
equivalent non-volatile storage system. In this format, they can be
transported easily and loaded via a host computer into system
memory for execution. Alternatively in smaller, self-contained
environments, such as personal digital assistants, the above
routines can be implemented as part of a non-volatile storage
portion of a stand-alone integrated circuit, or embedded as part of
a non-volatile storage structure in an architecture of a typical
microprocessor or RISC processor. The instructions for carrying out
such routines can be encoded or implemented in a silicon substrate
as is done with other program ROMs, and using conventional
manufacturing techniques.
[0050] Although the present invention has been described in terms
of a preferred embodiment, it will be apparent to those skilled in
the art that many alterations and modifications may be made to such
embodiments without departing from the teachings of the present
invention. Accordingly, it is intended that the all such
alterations and modifications be included within the scope and
spirit of the invention as defined by the appended claims.
* * * * *