Computer processes for detecting and correcting writing problems associated with nominalizations
First Claim
Patent Images
1. A text analysis method relating to nominalizations, the method comprising:
- ascertaining automatically whether or not at least one sign is present in text, the sign relating to a possible writing problem, the sign comprising a characteristic of nominalizations;
when the sign is present in the text, determining whether the sign is part of a word or word group that is present in a database of false positives; and
when it is determined that the sign is not part of a word or word group that is present in the database of false positives, determining an edit to propose to a user, the edit comprising changing the text to remove a nominalization;
wherein the method comprises comparing a selected sign with a first list of false positives and if the selected sign does not match a false positive, determining if the word preceding the selected sign is an adjective from an adjective pool and if a match is found, determining if the selected sign and the adjective match any entries in a second list of false positives, and if a match is found, ignoring the selected sign and the preceding adjective;
said method performed by a computer system that comprises one or more computers.
2 Assignments
0 Petitions
Accused Products
Abstract
Disclosed are systems and methods for analyzing and improving document readability. For example, a method of text analysis is disclosed that can deal with problems related nominalizations in text. Such a method can include the following steps: ascertaining automatically whether or not at least one sign is present in text, the sign relating to a possible writing problem, the sign comprising a characteristic of nominalizations; and proposing an edit to a user, the edit comprising changing a nominalization if the sign is determined to not be a false positive. Various other rules for improving text are also disclosed.
36 Citations
19 Claims
-
1. A text analysis method relating to nominalizations, the method comprising:
-
ascertaining automatically whether or not at least one sign is present in text, the sign relating to a possible writing problem, the sign comprising a characteristic of nominalizations; when the sign is present in the text, determining whether the sign is part of a word or word group that is present in a database of false positives; and when it is determined that the sign is not part of a word or word group that is present in the database of false positives, determining an edit to propose to a user, the edit comprising changing the text to remove a nominalization; wherein the method comprises comparing a selected sign with a first list of false positives and if the selected sign does not match a false positive, determining if the word preceding the selected sign is an adjective from an adjective pool and if a match is found, determining if the selected sign and the adjective match any entries in a second list of false positives, and if a match is found, ignoring the selected sign and the preceding adjective; said method performed by a computer system that comprises one or more computers. - View Dependent Claims (2, 3)
-
-
4. A text analysis method relating to nominalizations, the method comprising:
-
ascertaining automatically whether or not at least one sign is present in text, the sign relating to a possible writing problem, the sign comprising a characteristic of nominalizations; when the sign is present in the text, determining whether the sign is part of a word or word group that is present in a database of false positives; when it is determined that the sign is not part of a word or word group that is present in the database of false positives, determining an edit to propose to a user, the edit comprising changing the text to remove a nominalization; and comparing a selected sign with a first list of false positives and if the selected sign does not match a false positive, determining if the first four words preceding the selected sign include more than one of a “
to be”
verb and/or a verb from a verb pool;said method performed by a computer system that comprises one or more computers. - View Dependent Claims (5, 6, 7, 8, 9, 10, 11)
-
-
12. A text analysis method relating to nominalizations, the method comprising:
-
ascertaining automatically whether or not at least one sign is present in text, the sign relating to a possible writing problem, the sign comprising a characteristic of nominalizations; when the sign is present in the text, determining whether the sign is part of a word or word group that is present in a database of false positives; when it is determined that the sign is not part of a word or word group that is present in the database of false positives, determining an edit to propose to a user, the edit comprising changing the text to remove a nominalization; and searching for one or more of the following signs;
“
it”
appearing within the first six words preceding a nominalization, a nominalization modified by a possessive pronoun or noun, a prepositional phrase immediately following a nominalization, and a possessive noun or pronoun in the third person;said method performed by a computer system that comprises one or more computers.
-
-
13. A computer system for automatic text evaluation, the system comprising:
-
a rule in a database of rules, said rule having; detection logic for detecting whether a particular writing problem exists in text, the detection logic comprising at least one sign that indicates the possible occurrence or absence of the writing problem, the writing problem comprising a nominalization and the at least one sign comprising words ending in at least one of;
“
ion,”
“
ions,”
“
al,”
“
ant,”
“
ance,”
“
ancy,”
“
ent,”
“
enc,”
“
ency,”
“
ity,”
“
ing,”
“
sis,”
“
ise,” and
“
ure”
; andcorrection logic for correcting the problem by specifying a proposed edit, the correction logic comprising logic for; determining if the sign is part of a word or word group that is present in a database of nominalization false positives; and when the sign is not part of a word or word group that is present in the database of nominalization false positives, determining an edit to propose to a user, the edit comprising changing the text to remove the nominalization; a computer processor configured to apply the rules to text; and a user interface configured to communicate any proposed edit to the user; wherein the correction logic is capable of selecting a verb closest to a detected sign and deleting the selected verb along with an article and an occurrence of the term “
in”
between the selected verb and the detected sign. - View Dependent Claims (14, 15)
-
-
16. Non-transitory computer storage which stores executable code capable of causing a computing system to perform a process that comprises:
-
scanning a sentence for at least one sign of a possible writing problem relating to nominalization; in response to detecting an occurrence of said sign, determining whether the sign is part of a word or word group that is present in a database of false positives; and at least partly in response to determining that the sign is not part of a word or word group that is present in the database of false positives, determining an edit that can be made to the sentence to improve readability of the sentence, the edit comprising changing the sentence to remove a nominalization, wherein the edit comprises deleting a verb closest to the occurrence of the sign. - View Dependent Claims (17, 18, 19)
-
Specification