Parsing, analysis and scoring of document content

• US 8,065,307 B2
• Filed: 12/20/2006
• Issued: 11/22/2011
• Est. Priority Date: 12/20/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Haslam, Brian Dean, Menezes, Arul, Evans, Patrick Wayne John, Santos, Patrick
• Primary Examiner(s): Rones; Charles
• Assistant Examiner(s): Mahmood; Rezwanul
• Application Number:
  US11/613,958
• Publication Number:
  US 20080154848A1
• Time in Patent Office: 1,798 Days
• Field of Search: 707/3, 707/102, 707/736, 707/738, 707/750, 707/751, 707/755
• US Class Current: 707/738
• CPC Class Codes: G06F 16/382 using citations hypermedia G06F16/94
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 2216/11 Patent retrieval

Abstract

The present invention may be used to analyze subject content, search and analyze reference content, compare the subject and reference content for similarity, and output comparison reports between the subject and reference content. The present invention may incorporate and utilize text from intrinsic and/or extrinsic subject documents. The analysis may employ a variety of metrics, including scores generated from a natural language processing system, scores based on classification similarity, scores based on proximity similarity, and in the case of analysis of patent documents, scores based on measurement of claims.

Claims

1. A method in a computer to automatically analyze a dependent claim, the method in the computer comprising:

by the computer, parsing the dependent claim to obtain one or more claim strings, each claim string comprising words within a separate dependent claim element of the dependent claim and each claim string corresponding to the separate dependent claim element of the dependent claim;

by the computer, determining a number of claim strings;

by the computer, generating a number of claim strings score by merging the number of claim strings with a first factor;

initializing a dependent claim score to the number of claim strings score; and

for each claim string parsed from the dependent claim;

by the computer, computing a word count for the words within the claim string,

by the computer, determining a weighted word count based on the word count and a second factor,

by the computer, merging the weighted word count with the dependent claim score;

by the computer, determining a parent claim score associated with a parent claim of the dependent claim;

merging the parent claim score with the dependent claim score.

2. The method in the computer of claim 1, where parsing employs a natural language processing system, the natural language processing system determining parts of speech of claim words and logical relations between claim words, each logical relation comprising two or more words and a logical relationship between the two or more words;

and

in response to the parts of speech and logical relations associated with the claim words, determining claim strings partially based on the parts of speech and logical relations.

3. The method in the computer of claim 2, further comprising generating output data that comprises one or more rows, where each row comprises a dependent claim and the dependent claim score for the dependent claim, the rows ordered from broadest dependent claim to narrowest dependent claim according to the dependent claim score.

4. The method in the computer of claim 2, further comprising:

determining a broadest claim according to a minimum claim score; and

displaying the broadest claim.

5. The method in the computer of claim 1, further comprising:

determining a broadest claim according to a minimum claim score for a first document;

determining a broadest claim according to another minimum claim score for a second document;

generating a display including an identifier associated with the first document, an identifier associated with the second document, wherein the identifiers are ordered by minimum claim score.

6. The method in the computer of claim 1, wherein the first factor associated with the number of claim strings is about 2.0 and the second factor associated with the word count is about 0.25.

7. A method in a computer comprising:

automatically parsing a dependent claim to obtain one or more claim strings, each claim string comprising words within a dependent claim element;

automatically determining a number of claim strings within the dependent claim;

automatically generating a number of claim strings score by merging the number of claim strings with a first factor;

automatically initializing a dependent claim score to the number of claim strings score;

for each claim string parsed from the claim;

automatically computing a word count for the words within the claim string;

automatically determining a weighted word count based on the word count and a second factor;

automatically merging the weighted word count with the dependent claim score;

automatically determining a parent claim score associated with a parent claim of the dependent claim;

automatically merging the parent claim score with the dependent claim score;

displaying the dependent claim and the dependent claim score in a hierarchical display.

8. The method in the computer of claim 7, where parsing employs a natural language processing system, the natural language processing system determining logical relations between claim words, each logical relation comprising two or more words and a logical relationship between the two or more words.

9. The method in the computer of claim 7, further comprising generating output data that comprises one or more rows, where each row comprises a claim and the claim score for the claim, the rows ordered from broadest claim to narrowest claim according to the claim score.

10. The method in the computer of claim 7, further comprising:

determining a broadest claim according to a minimum claim score; and

displaying the broadest claim.

11. The method performed by the computer of claim 7, further comprising:

determining a broadest claim according to a minimum claim score for a first document;

determining a broadest claim according to another minimum claim score for a second document;

generating a display including an identifier associated with the first document, an identifier associated with the second document, wherein the identifiers are ordered by minimum claim score.

12. A system for performing claim analysis, comprising:

a computer, comprising;

an analysis application configured to perform steps, comprising;

parsing a dependent claim to obtain one or more claim strings, each claim string comprising words within a dependent claim element;

determining a number of claim strings of the dependent claim;

initializing a dependent claim score by merging the number of claim strings with a first factor; and

for each claim string parsed from the dependent claim;

computing a word count for the words within the claim string,

determining a weighted word count based on the word count and a second factor,

merging the weighted word count to the dependent claim score;

determining a parent claim score associated with a parent claim of the dependent claim; and

merging the parent claim score with the dependent claim score.

13. The system of claim 12, further comprising:

a natural language processing system wherein the natural language processing system determines logical relations within the claim, each logical relation comprising two or more words and a logical relationship between the two or more words, and the natural language processing system determines the claim strings based on logical relations.

14. The system of claim 12, the analysis application further configured to perform steps for:

generating output data that comprises one or more rows, where each row comprises a claim and the claim score for the claim, the rows ordered from broadest claim to narrowest claim according to the claim score.

15. The system of claim 12, further configured to perform steps for:

determining a broadest claim according to a minimum claim score; and

displaying the broadest claim.

16. The system of claim 12, further configured to perform steps for:

determining a broadest claim according to a minimum claim score for a first document;

determining a broadest claim according to another minimum claim score for a second document;

generating a display including an identifier associated with the first document, an identifier associated with the second document, wherein the identifiers are ordered by minimum claim score.

17. The system of claim 12, wherein the first factor associated with the number of claim strings is about 2.0 and the second factor associated with the word count is about 0.25.

18. The system of claim 17, wherein merging the number of claim strings with the first factor comprises multiplying the number of claim strings by about 2.0.

19. The system of claim 18, wherein determining the weighted word count comprises multiplying the word count by about 0.25.