Method for determining overall effectiveness of a document

• US 7,391,885 B2
• Filed: 07/28/2004
• Issued: 06/24/2008
• Est. Priority Date: 07/30/2003
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Jhong, Yoon J., Wenn, John C. II, Seo, Youngseok, Omega, Noel S., Orlov, Leonid, Gloger, Paul B., Harrington, Steven J., Manchala, Daniel W.
• Primary Examiner(s): Johns; Andrew W
• Application Number:
  US10/900,608
• Publication Number:
  US 20050071743A1
• Time in Patent Office: 1,427 Days
• Field of Search: 382/100, 382/112, 382/229, 382/286, 382/288, 382/291, 382/292, 382/317, 715/200, 715/205, 715/243, 715/244, 715/245, 715/249
• US Class Current: 382/112
• CPC Class Codes: G06F 16/382 using citations hypermedia G06F16/94

Abstract

The present invention is directed to a method for determining a document'"'"'s overall effectiveness or quality using a technique that employs detecting correlation between document citation rate and document presentation elements such as style and layout. A document'"'"'s citation rate is the number of citations of or references to that document from other documents. This is taken as an indicator of a document'"'"'s overall effectiveness. This invention employs automated means to obtain, for a sample of documents, both presentation data and citation rate data. Presentation data is obtained, for each document in the sample, by automated inspection of the document, for stylistic elements. The citation rate for each document is based on the number of citations (e.g., hyperlinks) to that document from another set of documents, the larger the set the better. The present invention then computes the statistical correlation of document citation rate versus presentation elements used, in a straightforward manner to identify correlation between the citation rate and presentation element(s).

Claims

1. A method for characterizing at least one target document'"'"'s overall effectiveness amongst a set of documents, comprising:

a) obtaining at least one indicator for said target document relating to its citation rate within the set of documents;

b) characterizing at least one common presentation element for each of said documents in the set of documents;

c) computing a statistical correlation between the indicator and the at least one common stylistic element for said target document and the set of documents; and

d) employing said correlation as an indicator of said target document'"'"'s overall effectiveness.

2. The method of claim 1, wherein the indicator for said target document relating to citation rate is the number of references made to the target document from other documents within the set of documents.

3. The method of claim 2 wherein said citation rate for a target document is the number of hyperlinks from all other documents within the set of documents to said target document.

4. The method of claim 1 wherein said presentation data is obtained by automated inspection of the document.

5. The method of claim 1 wherein said presentation data includes stylistic data.

6. The method of claim 1 wherein said presentation data includes layout data.

7. The method of claim 1 wherein the set of documents is selected from at least a portion of the documents available on a network.

8. The method of claim 7 wherein the network is the world wide web.

9. The method of claim 1 wherein the set of documents is limited to hyperlinked documents with machine-perceptible presentation elements.

10. A method for characterizing effectiveness of a document, comprising:

(a) measuring a predetermined set of characteristics of the document;

(b) quantizing the measured predetermined set of characteristics of the document; and

(c) generating a quantized interest value for the document based on a combining function that includes a citation-correlation aspect, the predetermined combining function combining the quantized measured predetermined set of characteristics, the quantized interest value being a measure of quality of the document.

11. The method of claim 10, wherein the citation-correlation aspect is a characterization of the overall effectiveness of at least one presentation element of a target document amongst a set of documents, said effectiveness being obtained by the method comprising:

a) obtaining at least one indicator for said target document relating to its citation rate within the set of documents;

b) characterizing at least one common presentation element for each of said documents in the set of documents;

c) computing a statistical correlation between the indicator and the at least one common stylistic element for said target document and the set of documents; and

d) employing said correlation as an indicator of said target document'"'"'s overall effectiveness.

12. The method of claim 11, wherein the overall effectiveness is a weighted product of the quantized measured predetermined set of characteristics, and where said weighting is a function of the citation-correlation aspect.