Method of searching text to find relevant content and presenting advertisements to users

• US 20100094840A1
• Filed: 10/14/2008
• Published: 04/15/2010
• Est. Priority Date: 03/30/2007
• Status: Active Grant

• Current Assignee: Needlebot, Inc.
• Original Assignee: Gan Deniz Akyuz, Stuart Donnelly
• Inventors: Akyuz, Gan Deniz, Donnelly, Stuart
• Granted Patent: US 8,271,476 B2
• Time in Patent Office: Days
• US Class Current: 707/705
• CPC Class Codes: G06F 16/3331 Query processing
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

A method of locating relevant documents wherein documents are given a fingerprint comprising weights associated with particular topic categories of a classification system, each weight representing a degree to which the document relates to the particular topic category. Documents whose fingerprints have a predetermined degree of mathematical overlap with the fingerprint may be considered relevant. Fingerprints of texts may be combined to define user fingerprints and user fingerprints may be combined to define community user fingerprints, which can be monitored. Changes in the interest of the online community may be monitored and automatically reported to advertisers who can then instantly adjust their advertising campaign. Similarly, changes in community interest may be reported to pharmaceutical companies monitoring drug side effects. The methods may be used in combination with keyword searching.

Claims

1. A method of measuring changes in interest of a community of users, comprising:

obtaining user fingerprints of users that constitute the community at a first time, the user fingerprints comprising vectors whose components are scalar cumulative weights associated with topic categories;

combining the vectors that comprise the user fingerprints to obtain a first community user fingerprint;

obtaining user fingerprints of users that constitute the community at a second time, the user fingerprints comprising vectors whose components are scalar cumulative weights associated with topic categories;

combining the vectors that comprise the user fingerprints obtained at the second time to obtain a second community user fingerprint; and

calculating a mathematical difference between the first community user fingerprint and the second community user fingerprint.

2. The method of claim 1, further comprising obtaining user fingerprints of users that constitute the community at a third time and at subsequent times, the user fingerprints comprising vectors whose components are scalar cumulative weights associated with topic categories;

combining the user fingerprints obtained at the third time to obtain a third community user fingerprint and combining the user fingerprints obtained at subsequent times to obtain subsequent community user fingerprints; and

calculating a community interest over time.

3. The method of claim 1, wherein the user fingerprints are combined by weighting users equally.

4. The method of claim 1, further comprising combining the user fingerprints to create first and second community user fingerprints by weighting the users by a number of searches conducted by the user.

5. The method of claim 1, further comprising combining the user fingerprints to create first and second community user fingerprints by weighting the users by a number of links obtained during searches conducted by the user.

6. The method of claim 1, further comprising calculating the mathematical difference by subtracting the first community user fingerprint from the second community user fingerprint to obtain a vector.

7. The method of claim 1, further comprising calculating the mathematical difference by using a distance formula for high dimensional space to measure a distance between an end of a first vector representing the first community user fingerprint and an end of a second vector representing the second community user fingerprint to obtain a scalar number.

8. The method of claim 1, further comprising calculating the mathematical difference by computing a difference vector running from an end of a first vector representing the first community user fingerprint to an end of a second vector representing the second community user fingerprint.

9. The method of claim 1, further comprising normalizing a vector representing a first or a second community user fingerprint by dividing each scalar cumulative weight in the vector by the square root of the sum of the squares of the scalar cumulative weight in the vector.

10. The method of claim 1, further comprising notifying an advertiser automatically when the mathematical difference reaches a threshold amount.

11. A method of detecting an upsurge in side effects of a drug, comprising:

obtaining user fingerprints of users that constitute an online community at a first time, the user fingerprints comprising vectors whose components are scalar cumulative weights associated with topic categories the components including scalar cumulative weights associated with a topic category relating to an increase in interest in use of particular drugs;

combining the vectors that comprise the user fingerprints to obtain a first community user fingerprint;

obtaining user fingerprints of users that constitute the community at a second time, the user fingerprints comprising vectors whose components are scalar cumulative weights associated with topic categories the components including scalar cumulative weights associated with a topic category relating to an increase in interest in use of particular drugs;

combining the vectors that comprise the user fingerprints obtained at the second time to obtain a second community user fingerprint; and

calculating a mathematical difference between the first community user fingerprint and the second community user fingerprint to detect an increase in interest in use of particular drugs.

12. The method of claim 11, further comprising limiting the user fingerprints forming a basis of the first and second community user fingerprints to user fingerprints that are based on fingerprints of documents, the documents containing a particular keyword.

13. The method of claim 11, further comprising calculating the mathematical difference by computing a difference vector running from an end of a first vector representing the first community user fingerprint to an end of a second vector representing the second community user fingerprint.

14. The method of claim 13, further comprising finding the specific documents that discuss the returns of unused prescriptions for drugs by calculating a mathematical overlap between the difference vector and the vectors of the documents that contributed to the difference vector.

15. A method of detecting an upsurge in side effects of a drug, comprising:

obtaining user fingerprints of users that constitute an online community at a first time, the user fingerprints comprising vectors whose components are scalar cumulative weights associated with topic categories the components including scalar cumulative weights associated with a topic category relating to reporting side effects of drugs;

combining the vectors that comprise the user fingerprints to obtain a first community user fingerprint;

obtaining user fingerprints of users that constitute the community at a second time, the user fingerprints comprising vectors whose components are scalar cumulative weights associated with topic categories the components including scalar cumulative weights associated with a topic category relating to reporting side effects of drugs;

combining the vectors that comprise the user fingerprints obtained at the second time to obtain a second community user fingerprint; and

calculating a mathematical difference between the first community user fingerprint and the second community user fingerprint to detect a change in a number of side effects for drugs reported.

16. The method of claim 15, further comprising limiting the user fingerprints forming a basis of the first and second community user fingerprints to user fingerprints that are based on fingerprints of documents, the documents containing a particular keyword identifying the drug.

17. The method of claim 15, further comprising calculating the mathematical difference by computing a difference vector running from an end of a first vector representing the first community user fingerprint to an end of a second vector representing the second community user fingerprint.

18. The method of claim 17, further comprising finding the specific documents that discuss the side effects of drugs by calculating a mathematical overlap between the difference vector and the vectors of the documents that contributed to the difference vector.