METHOD AND SYSTEM OF SCORING DOCUMENTS BASED ON ATTRIBUTES OBTAINED FROM A DIGITAL DOCUMENT BY EYE-TRACKING DATA ANALYSIS

• US 20150213123A1
• Filed: 01/27/2014
• Published: 07/30/2015
• Est. Priority Date: 08/29/2011
• Status: Active Grant

• Current Assignee: Amit Karmarkar, Richard R. Peters
• Original Assignee: Buckyball Mobile, Inc.
• Inventors: Karmarkar, Amit, Peters, Richard R
• Granted Patent: US 9,727,645 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/436 using biological or physiological data of a human being, e.g. blood pressure, facial expression, gestures
• CPC Class Codes: G06F 16/903 Querying for retrieval from the web G06F16/953
• CPC Class Codes: G06F 16/9035 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

In one exemplary embodiment, a set of attributes derived from an element of a first digital document is obtained. The element is identified from eye-tracking data of a user viewing the digital document. A search query of a database comprising at least one query term is received. A set of documents in the database is identified according to the search query. An attribute score is determined for each document. The set of documents are sorted according to the attribute score. Optionally, a commonality between the query term and at least one member of the set of attributes ma be determined. The search query may be generated by the user. The database may be a hypermedia database.

Claims

1. A computer-implemented method for scoring documents comprising:

obtaining a set of attributes derived from an element of a web page, wherein the element is identified from eye-tracking data of a user viewing the web page, wherein the element of the web page comprises an image portion of the web page, and wherein an attribute comprises an identity of the image portion of the web page;

weighting each attribute of the set of attributes based on the eye-tracking data with respect to each attribute and a time-stamp value of the eye-tracking data;

receiving a search query of a database, wherein the search query comprises at least one query term, wherein the database comprises an index of web pages;

identifying a set of web page documents in the database according to the search query;

determining an attribute score for each document based on a relevancy of each web page document to the set of weighted attributes; and

sorting the search query results according to the attribute score of each web page document.

2. The computer-implemented method of claim 1, further comprising:

determining a commonality between the query term and at least one member of the set of attributes.

3. The computer-implement method of claim 2, wherein determining the attribute score for each document comprises:

determining a number of matches or a probability weight of a match between the set of attributes and each document.

4. The computer-implement method of claim 2, wherein determining the attribute score for each document comprises:

determining a number of matches between the set of attributes and at least one theme of each document.

5. The computer-implement method of claim 2, wherein the weight of each attribute decreases as a function of time.

6. The computer-implemented method of claim 5, wherein the element of the digital document comprises a digital image.

7. The computer-implemented method of claim 5, wherein the element of the digital document comprises an image.

8. The computer-implemented method of claim 6, wherein attribute of the image comprises at least one of a color of the image or an object represented in the image.

9. The computer-implemented method of claim 2 further comprising:

obtaining a plurality of attributes from a plurality of digital documents over a specified period.

10. The computer-implemented method of claim of 9 further comprising:

scoring each attribute of the plurality of attributes based on a relevancy of each attribute to the user.

11. The computer-implemented method of claim of 10, wherein scoring each attribute of the plurality of attributes based on a relevancy of each attribute to the user further comprises:

determining a genre of each attribute of the plurality of attributes; and

scoring each genre based on a number of member attributes.

12. The computer-implemented method of claim of 10, wherein determining the attribute score for each document comprises:

determining a genre that is relevant to the search query; and

determining a number of matches between the attributes the relevant genre and the components of each document.

13. The computer-implemented method of claim of 1, wherein the web page documents comprise advertisement copy documents.

14. The computer-implemented method of claim of 1, wherein the database is a hypermedia database.

15. A non-transitory computer-readable storage medium comprising computer-executable instructions for scoring a set of documents, the computer-executable instructions comprising instructions for:

obtaining at least one attribute derived from an element of an augmented reality view, wherein the element is identified from eye-tracking data of a user viewing the an augmented reality view, wherein the element of the an augmented reality view comprises an image of a physical object obtained by a digital camera, and wherein an attribute comprises an identity of the an augmented reality view;

weighting each attribute of the set of attributes based on the eye-tracking data with respect to each attribute;

receiving a search query of a database, wherein the search query comprises at least one query term, and wherein the database comprises an index of documents;

identifying a set of documents in the database according to the search query;

determining an attribute score for each document based on a relevancy of each document to the set of weighted attributes; and

ranking the set of documents according to the attribute score of each document.

16. The non-transitory computer-readable storage medium of claim 15, wherein the an augmented reality view further comprises at least one of a web page, an image display of a digital camera or a multimedia messaging system message.

17. The non-transitory computer-readable storage medium of claim 15, wherein the attribute score of a document comprises a value determined by a comparison between a term vector representation of the document and the attribute.

18. The non-transitory computer-readable storage medium of claim 15 further comprising:

generating a database of attributes associated with a user viewing a plurality of digital documents over a specified period; and

ranking the database of attributes based on an attribute frequency value in the database.

19. The non-transitory computer-readable storage medium of claim 18, wherein the at least one attribute derived from the element of a digital document comprises at least a highest-ranked attribute.

20. A computer system for scoring a set of documents identified by a search query result, the system comprising:

memory configured to store the score of each document identified by a search query result; and

one or more processors configured to;

obtain at least one attribute derived from an element of a web page, wherein the element is identified from eye-tracking data of a user viewing the web page, wherein the element of the web page comprises a text portion or an image portion of the web page, and wherein an attribute comprises an identity of the image portion of the web page;

weight each attribute of the set of attributes based on the eye-tracking data with respect to each attribute;

receive a search query of a database, wherein the search query comprises at least one query term, and wherein the database comprises an index of documents;

identify a set of documents in the database according to the search query;

determine an attribute score for each document based on a relevancy of each document to the set of weighted attributes; and

rank the set of documents according to the attribute score of each document.