Methods and systems for analyzing incident reports

• US 20070005650A1
• Filed: 06/30/2005
• Published: 01/04/2007
• Est. Priority Date: 06/30/2005
• Status: Active Grant

• Current Assignee: The Boeing Co.
• Original Assignee: The Boeing Co.
• Inventors: Xue, Ping, Coen, Gary
• Granted Patent: US 7,613,667 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

Systems and methods for examining text, such as incident reports, are disclosed. In one embodiment, a method includes specifying a client profile including at least one scenario of interest, the scenario of interest being formulated in an event calculus formalism; analyzing a portion of text for relevant information at least partially described by the scenario of interest; detecting relevant information including detecting a positive match between at least part of information relevant to the scenario of interest and at least part of the portion of text being analyzed; upon detecting such relevant information, extracting the relevant information; representing the extracted relevant information in the event calculus formalism; and performing an inference process on the extracted relevant information represented in the event calculus formalism.

Claims

1. A method for automatically analyzing a portion of text such as that typically occurring in an incident report, comprising:

specifying a client profile including at least one scenario of interest, the scenario of interest being formulated in an event calculus formalism;

analyzing the portion of text for relevant information at least partially described by the scenario of interest;

detecting relevant information including detecting a positive match between at least part of information relevant to the scenario of interest and at least part of the portion of text being analyzed;

upon detecting such relevant information, extracting the relevant information;

representing the extracted relevant information in the event calculus formalism; and

performing an inference process on the extracted relevant information represented in the event calculus formalism.

2. The method of claim 1, wherein performing an inference process on the extracted relevant information in the event calculus formalism includes at least one of deductively determining when the extracted relevant information embodies the scenario of interest and abductively determining when the extracted relevant information embodies the scenario of interest.

3. The method of claim 2, wherein deductively determining when the extracted relevant information embodies the scenario of interest includes at least one of confirming the occurrence of the scenario of interest in the extracted relevant information and predicting the occurrence of the scenario of interest in the extracted relevant information.

4. The method of claim 3, wherein predicting the occurrence of the scenario of interest includes determining that the extracted relevant information only partially embodies the scenario of interest, and determining that a confidence weight attached to the partially-embodied scenario of interest is satisfied.

5. The method of claim 1, wherein performing an inference process includes abductively determining a most likely profiled scenario compatible with a plurality of user-specified events.

6. The method of claim 5, wherein abductively determining a most likely profiled scenario includes performing an inference process based on a logical formula represented as a conjunction F of a set of logical formulae describing events and activities drawn from the portion of text, the conjunction F including the plurality of user-specified events, and finding a logical description Δ

(M) of reported events such that F=Σ

(B)Σ

(E)Δ

(N)Δ

(M), where “

”

is the logical “

AND”

operator, and where;

Σ

(B) is a background theory comprising axioms for change in space-time, including changes in temporal and spatial locations;

Σ

(E) is a set of rules that relate a set of natural language representations with a set of event calculus representations; and

Δ

(N) is an axiom set comprising a logical description of client information requirements.

7. The method of claim 1, wherein performing an inference process includes determining that one or more confidence weights attached to a partially-embodied scenario of interest are satisfied.

8. The method of claim 1, wherein specifying a client profile having at least one scenario of interest includes specifying a client profile having at least one of an event and a state of affairs.

9. The method of claim 1, wherein specifying a client profile having at least one scenario of interest includes specifying a client profile having at most a confidence weight attached to the scenario of interest, the confidence weight identifying a minimum number of axioms that must be instantiated by the extracted relevant information in determining when the extracted relevant information embodies the scenario of interest.

10. The method of claim 1, wherein analyzing the portion of text for relevant information includes analyzing the portion of text using a natural language engineering component and providing a linguistic representation of at least part of the portion of text.

11. The method of claim 10, wherein representing the extracted relevant information in the event calculus formalism includes arranging the extracted information according to a formal set of truth conditions.

12. The method of claim 10, wherein representing the extracted relevant information in the event calculus formalism includes reducing a unit of information from the portion of text into a syntactic representation of noun and verb phrases, and subsequently reducing the syntactic representation into a semantic representation in the formalism of lexical-conceptual structures.

13. A computer program product for analyzing a portion of text such as that typically occurring in an incident report, comprising:

a first computer program portion adapted to specify a client profile including at least one scenario of interest, the scenario of interest being formulated in an event calculus formalism;

a second computer program portion adapted to analyze the portion of text for relevant information at least partially described by the scenario of interest;

a third computer program portion adapted to detect relevant information including detecting a positive match between at least part of information relevant to the scenario of interest and at least part of the portion of text being analyzed;

a fourth computer program portion adapted to, upon detecting such relevant information, extract the relevant information;

a fifth computer program portion adapted to represent the extracted relevant information in the event calculus formalism; and

a sixth computer program portion adapted to perform an inference process on the extracted relevant information represented in the event calculus formalism.

14. The computer program product of claim 13, wherein the sixth computer program portion is further adapted to perform an inference process including at least one of deductively determining when the extracted relevant information embodies the scenario of interest and abductively determining when the extracted relevant information embodies the scenario of interest.

15. The computer program product of claim 14, wherein the sixth computer program portion is further adapted to perform an inference process wherein deductively determining when the extracted relevant information embodies the scenario of interest includes at least one of confirming the occurrence of the scenario of interest in the extracted relevant information and predicting the occurrence of the scenario of interest in the extracted relevant information.

16. The computer program product of claim 15, wherein the sixth computer program portion is further adapted to perform an inference process, wherein predicting the occurrence of the scenario of interest includes determining that the extracted relevant information only partially embodies the scenario of interest, and determining that a confidence weight attached to the partially-embodied scenario of interest is satisfied.

17. The computer program product of claim 13, wherein the sixth computer program portion is further adapted to perform an inference process, wherein performing an inference process includes abductively determining a most likely profiled scenario compatible with a plurality of user-specified events.

18. The computer program product of claim 13, wherein the sixth computer program portion is further adapted to perform an inference process including determining that one or more confidence weights attached to a partially-embodied scenario of interest are satisfied.

19. The computer program product of claim 13, wherein the second computer program portion is further adapted to analyze the portion of text using a natural language engineering component and providing a linguistic representation of at least part of the portion of text.

20. A computer system for analyzing a portion of text such as that typically occurring in an incident report, comprising:

a memory component adapted to store at least a portion of the text; and

a processor coupled to the memory component, the processor comprising;

a first portion adapted to specify a client profile including at least one scenario of interest, the scenario of interest being formulated in an event calculus formalism;

a second portion adapted to analyze the portion of text for relevant information at least partially described by the scenario of interest;

a third portion adapted to detect relevant information including detecting a positive match between at least part of information relevant to the scenario of interest and at least part of the portion of text being analyzed;

a fourth portion adapted to, upon detecting such relevant information, extract the relevant information;

a fifth portion adapted to represent the extracted relevant information in the event calculus formalism; and

a sixth portion adapted to perform an inference process on the extracted relevant information represented in the event calculus formalism.

21. The computer system of claim 20, wherein the sixth portion is further adapted to perform an inference process including at least one of deductively determining when the extracted relevant information embodies the scenario of interest and abductively determining when the extracted relevant information embodies the scenario of interest.

22. The computer system of claim 21, wherein the sixth portion is further adapted to perform an inference process wherein deductively determining when the extracted relevant information embodies the scenario of interest includes at least one of confirming the occurrence of the scenario of interest in the extracted relevant information and predicting the occurrence of the scenario of interest in the extracted relevant information.

23. The computer system of claim 20, wherein the sixth portion is further adapted to perform an inference process, wherein performing an inference process includes abductively determining a most likely profiled scenario compatible with a plurality of user-specified events.

24. The computer system of claim 20, wherein the sixth portion is further adapted to perform an inference process including determining that one or more confidence weights attached to a partially-embodied scenario of interest are satisfied.

25. The computer system of claim 20, wherein the second portion is further adapted to analyze the portion of text using a natural language engineering component and providing a linguistic representation of at least part of the portion of text.