Systems and methods for natural language processing including morphological analysis, lemmatizing, spell checking and grammar checking

• US 8,762,130 B1
• Filed: 06/17/2009
• Issued: 06/24/2014
• Est. Priority Date: 06/17/2009
• Status: Active Grant

• Current Assignee: Softwin SRL Romania
• Original Assignee: Softwin SRL Romania
• Inventors: Diaconescu, Stefan Stelian, Paun, Bianca-Daniela, Dumitrascu, Ionut Mihai, Ingineru, Cristi Iulian, Bulibasa, Oana-Adriana, Rizea, Monica Mihaela
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): ADESANYA, OLUJIMI A
• Application Number:
  US12/486,546
• Time in Patent Office: 1,833 Days
• Field of Search: 704/7, 704/9, 704/10
• US Class Current: 704/9
• CPC Class Codes: G06F 40/20 Natural language analysis semantic analysis of natural language G06F40/30
• CPC Class Codes: G06F 40/232 Orthographic correction, e.g. spell checking or vowelisation
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique

Abstract

In some embodiments, a linguistic application exploits a linguistic knowledgebase (LKB) including, among others, lexicon data, inflection form data, and syntax data for a natural language such as English or Romanian. The application employs a set modules including a word retriever, a form generator, and a syntax checker, which are interconnected to perform a number of higher-level text-processing operations such as synthetic and analytic annotation, lemmatizing, spell checking, and grammar checking.

Claims

1. A system comprising at least one computer configured to form:

a linguistic knowledgebase (LKB) for a natural language, the LKB comprising a set of computer-readable lexicon declarations, a set of computer-readable inflected form declarations, and a set of computer-readable syntax rule declarations;

a computer-implemented word retriever connected to the LKB and configured to;

receive a first word,

perform a lookup of an inflected form declaration of the first word in the LKB, in response to performing the lookup of the inflected form declaration, perform a lookup of a lexicon declaration of the first word in the LKB,

determine a first word interpretation of the first word according to the lexicon declaration and the inflected form declaration, the first word interpretation comprising a lemma of the first word and an inflection indicator of the first word;

a computer-implemented form generator connected to the word retriever and configured to;

receive a second word not necessarily distinct from the first word,

produce a first set of words, each word of the first set of words having a predetermined spelling similarity to the second word, and

for each word of the first set of words, receive from the word retriever a second word interpretation of said each word of the first set of words;

a computer-implemented synthetic annotator connected to the word retriever and configured to;

receive a word sequence,

for each word of the word sequence, receive from the word retriever a third word interpretation of said each word of the word sequence, and

determine a synthetic annotation of the word sequence, the synthetic annotation comprising the third word interpretation of said each word of the word sequence; and

a computer-implemented syntax checker connected to the synthetic annotator and configured to;

receive the synthetic annotation from the synthetic annotator,

perform a lookup of a syntax rule declaration of the word sequence in the LKB according to the synthetic annotation, and

perform a syntactic analysis of the word sequence according to the syntax rule declaration, to determine a synthetic dependency tree of the word sequence.

2. The system of claim 1, further comprising a computer-implemented speller connected to the word retriever and form generator and configured to receive the first word interpretation from the word retriever;

determine according to the first word interpretation whether the LKB includes the inflected form declaration of the first word;

when the LKB does not include the inflected form declaration, receive from the form generator a second set of words having a predetermined spelling similarity to the first word; and

display at least a word of the second set of words to a user.

3. The system of claim 1, further comprising a computer-implemented lemmatizer connected to the word retriever and configured to

receive the first word interpretation from the word retriever;

and

display the lemma of the first word to a user.

4. The system of claim 1, further comprising a computer-implemented morphological analyzer connected to the synthetic annotator and syntax checker and configured to

receive the synthetic annotation from the synthetic annotator;

receive the synthetic dependency tree from the syntax checker;

determine an analytic annotation of the word sequence according to the synthetic annotation and the synthetic dependency tree, the analytic annotation comprising an analytic dependency tree of the word sequence, wherein a selected node of the analytic dependency tree comprises a plurality of nodes of the synthetic dependency tree; and

display to a user the analytic annotation or the analytic dependency tree.

5. The system of claim 1, further comprising a computer-implemented grammar checker connected to the syntax checker and configured to receive the synthetic dependency tree from the syntax checker;

and

display the synthetic dependency tree to a user.

6. The system of claim 1, wherein the LKB comprises Grammar Abstract Language (GAL) or Extensible Markup Language (XML) code.

7. A method comprising:

employing a computer-implemented word retriever to;

receive a first word,

perform a lookup of an inflected form declaration of the first word in a linguistic knowledgebase (LKB) of a natural language, the LKB comprising a set of computer-readable lexicon declarations, a set of computer-readable inflected form declarations, and a set of computer-readable syntax rule declarations,

in response to performing the lookup of the inflected form declaration, perform a lookup of a lexicon declaration of the first word in the LKB,

determine a first word interpretation of the first word according to the lexicon declaration and the inflected form declaration, the first word interpretation comprising a lemma of the first word and an inflection indicator of the first word;

employing a computer-implemented form generator connected to the word retriever to;

receive a second word not necessarily distinct from the first word,

produce a first set of words, each word of the first set of words having a predetermined spelling similarity to the second word, and

for each word of the first set of words, receive from the word retriever a second word interpretation of said each word of the first set of words;

employing a computer-implemented synthetic annotator connected to the word retriever to

receive a word sequence,

for each word of the word sequence, receive from the word retriever a third word interpretation of said each word of the word sequence, and

determine a synthetic annotation of the word sequence, the synthetic annotation comprising the third word interpretation of said each word of the word sequence; and

employing a computer-implemented syntax checker connected to the synthetic annotator and configured to

receive the synthetic annotation from the synthetic annotator,

perform a lookup of a syntax rule declaration of the word sequence in the LKB according to the synthetic annotation, and

perform a syntactic analysis of the word sequence according to the syntax rule declaration, to determine a synthetic dependency tree of the word sequence.

8. The method of claim 7, further comprising, in response to determining the first set of words, displaying at least a word of the first set of words on a computer-implemented display device.

9. The method of claim 7, further comprising, in response to performing the lookup of the lexicon declaration of the first word, displaying the lemma of the first word on a computer-implemented display device.

10. The method of claim 7, further comprising, in response to determining the synthetic dependency tree, displaying the synthetic dependency tree on a computer-implemented display device.

11. The method of claim 7, further comprising, in response to determining the synthetic dependency tree, determining an analytic dependency tree of the word sequence according to the synthetic dependency tree, wherein a selected node of the analytic dependency tree comprises a plurality of nodes of the synthetic dependency tree.

12. The method of claim 11, further comprising, in response to determining analytic dependency tree, displaying the analytic dependency tree on a computer-implemented display device.

13. The method of claim 7, wherein the LKB comprises Grammar Abstract Language (GAL) or Extensible Markup Language (XML) code.

14. A non-transitory computer-readable medium encoding instructions which, when executed by a computer system comprising at least one computer, cause the computer system to form:

a linguistic knowledgebase (LKB) for a natural language, the LKB comprising a set of computer-readable lexicon declarations, a set of computer-readable inflected form declarations, and a set of computer-readable syntax rule declarations;

a computer-implemented word retriever connected to the LKB and configured to;

receive a first word,

perform a lookup of an inflected form declaration of the first word in the LKB,

in response to performing the lookup of the inflected form declaration, perform a lookup of a lexicon declaration of the first word in the LKB,

determine a first word interpretation of the first word according to the lexicon declaration and the inflected form declaration, the first word interpretation comprising a lemma of the first word and an inflection indicator of the first word;

a computer-implemented form generator connected to the word retriever and configured to;

receive a second word not necessarily distinct from the first word,

produce a first set of words, each word of the first set of words being a spelling variant of the second word having a predetermined spelling similarity to the second word, and

for each word of the first set of words, receive from the word retriever a second word interpretation of said each word of the first set of words;

a computer-implemented synthetic annotator connected to the word retriever and configured to;

receive a word sequence,

for each word of the word sequence, receive from the word retriever a third word interpretation of said each word of the word sequence, and

determine a synthetic annotation of the word sequence, the synthetic annotation comprising the third word interpretation of said each word of the word sequence; and

a computer-implemented syntax checker connected to the synthetic annotator and configured to;

receive the synthetic annotation from the synthetic annotator,

perform a lookup of a syntax rule declaration of the word sequence in the LKB according to the synthetic annotation, and

perform a syntactic analysis of the word sequence according to the syntax rule declaration, to determine a synthetic dependency tree of the word sequence.

15. The non-transitory computer-readable medium of claim 14, wherein the instructions further cause the computer system to form a computer-implemented speller connected to the word retriever and form generator and configured to receive the first word interpretation from the word retriever;

determine according to the first word interpretation whether the LKB includes the inflected form declaration of the first word;

when the LKB does not include the inflected form declaration, receive from the form generator a second set of words having a predetermined spelling similarity to the first word; and

display at least a word of the second set of words to a user.

16. The non-transitory computer-readable medium of claim 14, wherein the instructions further cause the computer system to form a computer-implemented lemmatizer connected to the word retriever and configured to

receive the first word interpretation from the word retriever;

and

display the lemma of the first word to a user.

17. The non-transitory computer-readable medium of claim 14, wherein the instructions further cause the computer system to form a computer-implemented morphological analyzer connected to the synthetic annotator and syntax checker and configured to

receive the synthetic annotation from the synthetic annotator;

receive the synthetic dependency tree from the syntax checker;

determine an analytic annotation of the word sequence according to the synthetic annotation and the synthetic dependency tree, the analytic annotation comprising an analytic dependency tree of the word sequence, wherein a selected node of the analytic dependency tree comprises a plurality of nodes of the synthetic dependency tree; and

display to a user the analytic annotation or the analytic dependency tree.

18. The non-transitory computer-readable medium of claim 14, wherein the instructions further cause the computer system to form a computer-implemented grammar checker connected to the syntax checker and configured to

receive the synthetic dependency tree from the syntax checker;

and

display the synthetic dependency tree to a user.

19. The non-transitory computer-readable medium of claim 14, wherein the LKB comprises Grammar Abstract Language (GAL) or Extensible Markup Language (XML) code.