Functioning of a computing device by a natural language processing method comprising analysis of sentences by clause types

• US 9,779,083 B2
• Filed: 02/05/2016
• Issued: 10/03/2017
• Est. Priority Date: 09/18/2007
• Status: Active Grant

• Current Assignee: Elsevier Incorporated (RELX PLC)
• Original Assignee: Elsevier Incorporated (RELX PLC)
• Inventors: Daraselia, Nikolai
• Primary Examiner(s): Kazeminezhad, Farzad
• Application Number:
  US15/017,178
• Publication Number:
  US 20160154790A1
• Time in Patent Office: 606 Days
• Field of Search: 704 9, 704 8, 704 4
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/55 Rule-based translation
• CPC Class Codes: G06F 9/454 Multi-language systems; Localisation; Internationalisation

Abstract

Systems and methods for converting a natural language sentence into a computer-readable primitive sentence and extracting information therefrom are disclosed. A method includes identifying, by a processing device, a verbal block in the natural language sentence, splitting, by the processing device, the natural language sentence into one or more logical clauses, determining, by the processing device, a type for each logical clause, where the type indicates whether each logical clause contains an ambiguous verbal block, disambiguating, by the processing device, the ambiguous verbal block within each logical clause, where each verbal block is considered independently of a noun phrase, and constructing, by the processing device, the computer-readable primitive sentence for each ambiguous verbal block by duplicating a shared noun phrase of the ambiguous verbal block. The computer-readable primitive sentence improves functioning of a computing device by allowing the computing device to process the natural language sentence to obtain information therefrom.

Claims

1. A method for converting a natural language sentence into a computer-readable primitive sentence, the method comprising:

identifying, by a processing device, a verbal block in the natural language sentence;

splitting, by the processing device, the natural language sentence into one or more logical clauses;

determining, by the processing device, a type for each of the one or more logical clauses, wherein the type indicates whether each of the one or more logical clauses contains an ambiguous verbal block;

disambiguating, by the processing device, the ambiguous verbal block within each of the one or more logical clauses, wherein each verbal block is considered independently of a noun phrase; and

constructing, by the processing device, the computer-readable primitive sentence for each ambiguous verbal block by duplicating a shared noun phrase of the ambiguous verbal block,

wherein the computer-readable primitive sentence improves a functioning of a computing device by not requiring voluminous storage space in allowing the computing device to process the natural language sentence to obtain information from the natural language sentence.

2. The method of claim 1, further comprising:

extracting, by the processing device, the information from the computer-readable primitive sentence.

3. The method of claim 2, wherein extracting the information from the computer-readable primitive sentence comprises detecting sequences of words in the computer-readable primitive sentence using regular expression-like word patterns.

4. The method of claim 1, wherein determining the type comprises identifying the ambiguous verbal block by using a finite state automaton operating on syntactic categories of sentence lexemes.

5. The method of claim 4, wherein the finite state automaton is compiled from a set of regular expressions.

6. The method of claim 1, wherein each of the one or more logical clauses is a subordinate logical clause or a coordinate logical clause.

7. The method of claim 1, wherein each of the one or more logical clauses is one or more of a main clause, a which/that clause, an apposition clause, a V-complement clause, a prepositional modifier clause, an adverbial modifier clause, a conditional clause, a To-clause, an As-clause, a By-clause, and an ING-clause.

8. The method of claim 1, wherein disambiguating the ambiguous verbal block comprises:

examining, by the processing device, a first lexeme immediately preceding the ambiguous verbal block and a second lexeme immediately following the ambiguous verbal block;

matching, by the processing device, the first lexeme and the second lexeme to a rule to determine an ambiguous category; and

replacing the ambiguous category with a non-ambiguous category.

9. The method of claim 1, wherein disambiguating the ambiguous verbal block comprises performing a global disambiguating task.

10. A method for extracting information from a natural language sentence, the method comprising:

converting, by a processing device, the natural language sentence into a computer readable primitive sentence by detecting one or more logical clauses in the natural language sentence,

wherein the computer-readable primitive sentence improves a functioning of a computing device by not requiring voluminous storage space in allowing the computing device to process the natural language sentence to obtain information from the natural language sentence;

determining, by the processing device, a type for each of the one or more logical clauses, wherein the type indicates whether each of the one or more logical clauses contains an ambiguous verbal block; and

extracting, by the processing device, information using from the computer-readable primitive sentence based on the type for each of the one or more logical clauses.

11. The method of claim 10, wherein determining the type comprises identifying the ambiguous verbal block by using a finite state automaton operating on syntactic categories of sentence lexemes.

12. The method of claim 11, wherein the finite state automaton is compiled from a set of regular expressions.

13. The method of claim 10, wherein extracting the information from the computer-readable primitive sentence comprises detecting sequences of words in the computer-readable primitive sentence using regular expression-like word patterns.

14. The method of claim 10, wherein each of the one or more logical clauses is a subordinate logical clause or a coordinate logical clause.

15. The method of claim 10, wherein each of the one or more logical clauses is one or more of a main clause, a which/that clause, an apposition clause, a V-complement clause, a prepositional modifier clause, an adverbial modifier clause, a conditional clause, a To-clause, an As-clause, a By-clause, and an ING-clause.

16. A system for converting a natural language sentence into a computer-readable primitive sentence, the system comprising:

a processing device; and

a non-transitory, processor-readable storage medium comprising one or more programming instructions that, when executed, cause the processing device to;

identify a verbal block in the natural language sentence;

split the natural language sentence into one or more logical clauses;

determine a type for each of the one or more logical clauses, wherein the type indicates whether each of the one or more logical clauses contains an ambiguous verbal block;

disambiguate the ambiguous verbal block within each of the one or more logical clauses, wherein each verbal block is considered independently of a noun phrase; and

construct the computer-readable primitive sentence for each ambiguous verbal block by duplicating a shared noun phrase of the ambiguous verbal block,

wherein the computer-readable primitive sentence improves a functioning of a computing device by not requiring voluminous storage space in allowing the computing device to process the natural language sentence to obtain information from the natural language sentence.

17. The system of claim 16, wherein the non-transitory, processor-readable storage medium further comprises one or more programming instructions that, when executed, cause the processing device to:

extract the information from the computer-readable primitive sentence.

18. The system of claim 17, wherein the one or more programming instructions that, when executed, cause the processing device to extract the information from the computer-readable primitive sentence further comprise causing the processing device to detect sequences of words in the computer-readable primitive sentence using regular expression-like word patterns.

19. The system of claim 16, wherein the one or more programming instructions that, when executed, cause the processing device to determine the type further comprise causing the processing device to identify the ambiguous verbal block by using a finite state automaton operating on syntactic categories of sentence lexemes.

20. The system of claim 16, wherein the one or more programming instructions that, when executed, cause the processing device to disambiguate the ambiguous verbal block further comprise causing the processing device to perform at least one of a local disambiguating task and a global disambiguating task.