Systems and methods for using statistical techniques to reason with noisy data

• US 20050149230A1
• Filed: 01/06/2004
• Published: 07/07/2005
• Est. Priority Date: 01/06/2004
• Status: Active Grant

• Current Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Original Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Inventors: Gupta, Rakesh, Kochenderfer, Mykel J.
• Granted Patent: US 7,299,110 B2
• Time in Patent Office: Days
• US Class Current: 700/246
• CPC Class Codes: G06F 40/237 Lexical tools
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

Systems and methods are presented that enable logical reasoning even in the Presence of noisy (inconsistent) data. The knowledge base is processed in order to make it consistent and is also compiled. This processing includes checking and correcting spelling, removing stopwords, performing, grouping words of similar and related meaning, and compacting the knowledge base. A robot can use the processed knowledge base to perform many different types of tasks, such as answering a query, determining a course of action that is designed to achieve a particular goal, and determining its own location.

Claims

1. A method for processing a knowledge base, the method comprising the step of determining a set of relevant implications based on statistical methods.

2. The method of claim 1, further comprising the step of:

generalizing data in the knowledge base by replacing an implication associated with a first object with an implication associated with a second object, the second object being a hypernym of the first object.

3. The method of claim 1, further comprising the step of:

inferencing from the knowledge base.

4. The method of claim 3 wherein inferencing from the knowledge base comprises the step of:

inferencing from an object not present in the knowledge base.

5. The method of claim 4 wherein the object not present in the knowledge base is one of a hypernym and a synonym of an object that is present in the knowledge base.

6. The method of claim 3, further comprising the step of one of answering a query, determining how to perform a task, and determining how to fulfill a desire.

7. The method of claim 3, further comprising the step of one of determining consensus knowledge and determining common sense knowledge.

8. The method of claim 3, further comprising the step of determining a location of a robot.

9. The method of claim 1, further comprising the step of:

correcting a spelling of a word in the knowledge base, responsive to a determination that the spelling is incorrect.

10. The method of claim 1, further comprising the step of:

removing a word, responsive to a determination that the word is on a list of stopwords.

11. The method of claim 1, further comprising the step of:

replacing a word with a root form of the word, responsive to a determination that the word is not in root form.

12. The method of claim 1 wherein statistical methods comprise one of latent semantic analysis, spectral clustering, and principal component analysis.

13. An apparatus for processing a knowledge base, the apparatus comprising:

a statistical methods module, configured to determine a set of relevant implications based on statistical methods.

14. The apparatus of claim 13, further comprising:

a generalization module, configured to generalize data in the knowledge base by replacing an implication associated with a first object with an implication associated with a second object, the second object being a hypernym of the first object.

15. The apparatus of claim 13, further comprising:

an inference module, configured to inference from the knowledge base.

16. The apparatus of claim 15 wherein inferencing from the knowledge base comprises:

inferencing from an object not present in the knowledge base.

17. The apparatus of claim 16 wherein the object not present in the knowledge base is one of a hypernym and a synonym of an object that is present in the knowledge base.

18. The apparatus of claim 15, further comprising one of answering a query, determining how to perform a task, and determining how to fulfill a desire.

19. The apparatus of claim 15, further comprising one of determining consensus knowledge and determining common sense knowledge.

20. The apparatus of claim 15, further comprising determining a location of a robot.

21. The apparatus of claim 13, further comprising:

correcting a spelling of a word in the knowledge base, responsive to a determination that the spelling is incorrect.

22. The apparatus of claim 13, further comprising:

removing a word, responsive to a determination that the word is on a list of stopwords.

23. The apparatus of claim 13, further comprising:

replacing a word with a root form of the word, responsive to a determination that the word is not in root form.

24. The apparatus of claim 13 wherein statistical methods comprise one of latent semantic analysis, spectral clustering, and principal component analysis.

25. An apparatus for processing a knowledge base, the apparatus comprising:

means for determining a set of relevant implications based on statistical methods.

26. The apparatus of claim 25, further comprising:

means for generalizing data in the knowledge base by replacing an implication associated with a first object with an implication associated with a second object, the second object being a hypernym of the first object.

27. The apparatus of claim 25, further comprising:

means for inferencing from the knowledge base.

28. The apparatus of claim 27 wherein inferencing from the knowledge base comprises:

inferencing from an object not present in the knowledge base.

29. The apparatus of claim 28 wherein the object not present in the knowledge base is one of a hypernym and a synonym of an object that is present in the knowledge base.

30. The apparatus of claim 27, further comprising one of means for answering a query, means for determining how to perform a task, and means for determining how to fulfill a desire.

31. The apparatus of claim 27, further comprising one of means for determining consensus knowledge and means for determining common sense knowledge.

32. The apparatus of claim 27, further comprising means for determining a location of a robot.

33. The apparatus of claim 25, further comprising:

means for correcting a spelling of a word in the knowledge base, responsive to a determination that the spelling is incorrect.

34. The apparatus of claim 25, further comprising:

means for removing a word, responsive to a determination that the word is on a list of stopwords.

35. The apparatus of claim 25, further comprising:

means for replacing a word with a root form of the word, responsive to a determination that the word is not in root form.

36. The apparatus of claim 25 wherein statistical methods comprise one of latent semantic analysis, spectral clustering, and principal component analysis.