Linguistic Analysis

• US 20100030553A1
• Filed: 12/21/2007
• Published: 02/04/2010
• Est. Priority Date: 01/04/2007
• Status: Active Grant

• Current Assignee: Thinking Solutions Pty., Ltd.
• Original Assignee: Thinking Solutions Pty., Ltd.
• Inventors: Ball, John
• Granted Patent: US 8,600,736 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 40/216 using statistical methods
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/42 Data-driven translation
• CPC Class Codes: G06F 40/55 Rule-based translation
• CPC Class Codes: G06F 40/56 Natural language generation

Abstract

A method of operating a computer to perform linguistic analysis includes the steps of splitting an input text into words and sentences; for each sentence, comparing phrases in the sentence with known phrases stored in a database, as follows: for each word in the sentence, comparing its value and values of words following it with values of words of stored phrases, starting with the longest stored phrase that starts with that word, and working from longest to shortest; in the event a match is found for two or more consecutive words, and considering the words around the phrase, labelling the matched phrase with an overphrase that describes the grammar use of the matched phrase; after the penultimate word has been compared, recasting the sentence by replacing the matched phrases by their respective overphrases; and then repeating the comparison process with the recast sentence until there is no further recasting.

Claims

1. A method of operating a computer to perform linguistic analysis, comprising the steps of:

splitting an input text into words and sentences;

for each sentence, comparing phrases in each sentence with known phrases stored in a database, as follows;

for each word in the sentence, comparing a value thereof and values of the words following it with values of words of stored phrases, starting with a longest stored phrase that starts with that word, and working from longest to shortest;

in the event a match is found between the value of at least two consecutive words and the value of words of a stored phrase, labelling the matched at least two consecutive words with an overphrase that describes the matched value;

after a penultimate word has been compared, recasting the sentence by replacing matched phrases by respective overphrases; and

then repeating the step of comparing phrases with the recast sentence until there is no further recasting.

2. A method according to claim 1, wherein the value used in the step of comparing phrases corresponds to a literal meaning of the words.

3. A method according to claim 1, wherein the value used in the step of comparing phrases corresponds to a grammatical use of the words.

4. A method according to claim 1, wherein the value used in the step of comparing phrases corresponds to an attribute of the words.

5. A method according to claim 1,

wherein in a first comparison in the step of comparing a value, the value corresponds to one of:

a literal meaning,

a grammatical use and

an attribute, and

wherein in a second comparison in the step of comparing a value, the value is either of the two remaining values.

6. A method according to claim 5, wherein in a third comparison in the step of comparing a value, the value is the remaining value.

7. A method according to claim 1, wherein the step of comparing phrases comprises more than the comparison of the values of words.

8. A method according to claim 7, wherein the step of comparing phrases includes a conditional requirement dependent on words around the phrase.

9. A method according to claim 1, wherein the step of comparing phrases is performed according to a predetermined order.

10. A method according to claim 1, wherein phrases and words in the input text are converted to their overphrases according to a predetermined conversion order associated with the phrases and words.

11. A method according to claim 1, wherein the step of comparing further comprises an error correction step that corrects a misspelled word in the input sentence when a spelling error is encountered and repeats the step of comparing phrases with the corrected word.

12. A method according to claim 1, further comprising a setup step where words and their relationship with literal phrases, grammatical phrases and other attributes are stored in a database.

13. A method according to claim 1, further comprising the step of learning new words during operation.

14. A method according to claim 1, wherein word sense disambiguation is addressed by loading all word senses for each word and deselecting those senses that are not valid in the text provided.

15. A method according to claim 1, wherein word sense disambiguation is addressed by determining a consistent alternative for a word or phrase using grammar uses and attributes of the word or phrase.

16. A method according to claim 15, wherein equivalent phrases and words are in a different language.

17. A method according to claim 15, wherein to generate a text from grammar uses and attributes of words and phrases, the method further comprises the steps of:

for each overphrase in the input text, starting from a last overphrase determined, performing language generation to find an equivalent for the phrase or words represented by the overphrase, as follows;

obtaining all equivalent phrases of the overphrase from the database; and

if equivalent phrases are found in a targeted language, determining a word or phrase with equivalent attributes for the overphrase;

after the last overphrase has been considered, recasting the sentence by replacing the overphrases by respective equivalents, which may be overphrases at a lower level; and

then repeating the generation process until all overphrases are matched to words or their equivalents in the targeted language.

18. A method according to claim 1, implemented using an object-oriented (OO) programming language such that relationships between a word and related grammar use words which are words describing grammatical and semantic usage of the word;

phrases;

word senses; and

attributes are encapsulated within a word object.

19. A method according to claim 1, implemented in computer-based devices selected from the group consisting of servers, personal computers, laptops, mobile phones and personal digital assistants (PDAs).

20. A computer system which implements the method according to claim 1.

21. Software for programming a computer to perform the method according to claim 1.