System and method for extracting ontological information from a body of text

• US 9,244,909 B2
• Filed: 12/10/2012
• Issued: 01/26/2016
• Est. Priority Date: 12/10/2012
• Status: Active Grant

• Current Assignee: GE Digital Holdings LLC (GE Vernova, Inc.)
• Original Assignee: General Electric Company
• Inventors: Ng Tari, Luis Babaji, Moitra, Abha
• Primary Examiner(s): Opsasnick, Michael N
• Application Number:
  US13/710,218
• Publication Number:
  US 20140163955A1
• Time in Patent Office: 1,142 Days
• Field of Search: 704 1- 10
• US Class Current: 1/1
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

A system for extracting ontological information from a body of text is disclosed. The system parses one or more sentences from the body of text into parse tree format to generate a set of parsed sentences. The system further performs named-entity-recognition by identifying a subset of parsed sentences from the set of parsed sentences. A subset of noun phrases from the subset of parsed sentences are identified and the noun phrases are examined to classify the noun phrases as an entity or as a property. The system also identifies and outputs a conceptual relationship between the entity and the property based at least partially on grammatical relationship of the entity and the property.

Claims

1. A system for extracting ontological information from a body of text, said system comprising:

an input module configured to receive a verb phrase;

a parsing module configured to parse one or more sentences from the body of text into parse tree format, thereby generating a set of parsed sentences;

a named-entity-recognition module configured to;

identify a subset of parsed sentences from the set of parsed sentences based at least partially on the occurrence of the verb phrase within at least one parsed sentence of the set of parsed sentences;

identify a subset of noun phrases from the subset of parsed sentences based at least partially on grammatical relationship of each noun phrase of the subset of parsed sentences to the verb phrase;

classify a first noun phrase in the subset of noun phrases as an entity, thereby defining a first entity; and

classify a second noun phrase in the subset of noun phrases as a property, thereby defining a first property; and

a concept-extraction module configured to identify and output a conceptual relationship between the first entity and the first property based at least partially on grammatical relationship of the first entity and the first property within a first sentence, the output being to a display device for viewing by a user.

2. The system in accordance with claim 1, further comprising a parse tree database, wherein the parsing module is further configured to store the set of parsed sentences in the parse tree database.

3. The system in accordance with claim 1, further comprising a normalizing module configured to normalize one or more noun phrases in the subset of noun phrases by one or more of stemming, dropping extraneous articles, and dropping adjectives.

4. The system in accordance with claim 1, further comprising an aliasing module configured to alias one or more noun phrases in the subset of noun phrases by re-naming like noun phrases with a single alias.

5. The system in accordance with claim 1, wherein said named-entity-recognition module further configured to classify a third noun phrase as a second entity based at least partially on grammatical relationship of the third noun phrase to both a preposition and the first property in a second sentence.

6. The system in accordance with claim 1, wherein said named-entity-recognition module further configured to classify a third noun phrase as a second property based at least partially on grammatical relationship of the third noun phrase to both a preposition and the first entity in a second sentence.

7. The system in accordance with claim 1, wherein said named-entity-recognition module further configured to classify a first noun phrase as a first entity based at least partially on grammatical relationship of the first noun phrase to one or more prepositions within a second sentence.

8. The system in accordance with claim 1, wherein said named-entity-recognition module further configured to classify a second noun phrase as a first property based at least partially on grammatical relationship of the second noun phrase to one or more prepositions within a second sentence.

9. One or more non-transitory computer-readable storage media having computer-executable instructions embodied thereon, wherein when executed by at least one processor, the computer-executable instructions cause the processor to perform operations comprising:

converting one or more sentences of a body of text into parse tree format, thereby generating a set of parsed sentences;

identifying a verb phrase;

identifying a subset of parsed sentences from the set of parsed sentences based at least partially on the occurrence of the verb phrase within at least one parsed sentence of the set of parsed sentences;

identifying a subset of noun phrases from the subset of parsed sentences based at least partially on grammatical relationship of each noun phrase of the subset of parsed sentences to the verb phrase;

classifying a first noun phrase and a second noun phrase in the subset of noun phrases as one of an entity and a property, thereby defining one of a first entity and a first property;

identifying a conceptual relationship between the first entity and the first property based at least in part on grammatical relationship of the first entity and the first property within a first sentence; and

outputting the conceptual relationship as a relation between the first entity and the first property to an output device for viewing.

10. The non-transitory computer-readable storage media in accordance with claim 9, wherein the computer-executable instructions further cause said processor to perform operations comprising:

normalizing one or more noun phrases in the subset of noun phrases by one or more of stemming, dropping extraneous articles, and dropping adjectives; and

aliasing one or more noun phrases in the subset of noun phrases by renaming like noun phrases with a single alias.

11. The non-transitory computer-readable storage media in accordance with claim 9, wherein said computer-executable instructions further cause said processor to perform operations comprising classifying a second noun phrase as an entity based at least partially on grammatical relationship of the second noun phrase to both a preposition and a property in a second sentence.

12. The non-transitory computer-readable storage media in accordance with claim 9, wherein said computer-executable instructions further cause said processor to perform operations comprising classifying a second noun phrase as a property based at least partially on grammatical relationship of the second noun phrase to both a preposition and an entity in a second sentence.

13. The non-transitory computer-readable storage media in accordance with claim 9, wherein said computer-executable instructions further cause said processor to perform operations comprising classifying a first noun phrase in the subset of noun phrases as one of an entity and a property based at least partially on grammatical relationship of the first noun phrase to one or more prepositions within a second sentence.

14. A method for extracting ontological information from a body of text, said method implemented by at least one computer device including at least one processor and at least one memory device coupled to the at least one processor, said method comprising:

converting, by the at least one computer device, one or more sentences in the body of text into parse tree format, thereby generating a set of parsed sentences in the at least one memory device;

identifying a verb phrase;

identifying a subset of parsed sentences from the set of parsed sentences based at least partially on the occurrence of the verb phrase within at least one parsed sentence of the set of parsed sentences;

identifying a subset of noun phrases from the subset of parsed sentences based at least partially on grammatical relationship of each noun phrase of the subset of parsed sentences to the verb phrase;

classifying a first noun phrase and a second noun phrase in the subset of noun phrases as one of an entity and a property, thereby defining one of a first entity and a first property;

identifying a conceptual relationship between the first entity and the first property based at least in part on grammatical relationship of the first entity and the first property within a first sentence; and

outputting the conceptual relationship as an identified relation between the first entity and the first property to a presentation interface for viewing.

15. The method in accordance with claim 14, further comprising normalizing one or more of the noun phrases of the subset of noun phrases, wherein normalizing comprises at least one of stemming, dropping extraneous articles, and dropping adjectives.

16. The method in accordance with claim 14, further comprising aliasing one or more noun phrases of the subset of noun phrases comprising renaming like noun phrases with a single alias.

17. The method in accordance with claim 14, further comprising classifying a second noun phrase as an entity based at least partially on grammatical relationship of the second noun phrase to both a preposition and a property in a second sentence.

18. The method in accordance with claim 14, further comprising classifying a second noun phrase as a property based at least partially on grammatical relationship of the second noun phrase to both a preposition and an entity in a second sentence.

19. The method in accordance with claim 14, further comprising identifying a second verb phrase based at least partially on grammatical relationship of the second verb phrase to at least one of the first entity and the first property.

20. The method in accordance with claim 14, wherein classifying a first noun phrase is based at least partially on grammatical relationship of the first noun phrase to one or more prepositions within a second sentence.