Method, apparatus and article for displaying targeted content on web pages by predicting the group membership of individual visitors

• US 20030212594A1
• Filed: 05/03/2002
• Published: 11/13/2003
• Est. Priority Date: 04/22/2002
• Status: Active Grant

• Current Assignee: NBC Universal Media LLC (Comcast Corporation)
• Original Assignee: NBC Universal Media LLC (Comcast Corporation)
• Inventors: Hogan, Daniel Edward
• Granted Patent: US 7,783,514 B2
• Time in Patent Office: Days
• US Class Current: 705/14
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/0251 Targeted advertisements
• CPC Class Codes: G06Q 30/0277 Online advertisement

Abstract

A method, apparatus and article for displaying targeted content on web pages by predicting the group membership of individual visitors based on data items relating to a web browser request for a web page. The method includes applying the data items to a set of transfer functions to obtain a series of confidence factors relating to a plurality of predefined visitor groups, selecting one of the plurality of visitor groups based on the series of confidence factors, selecting content targeted to the selected group, and delivering the selected targeted content to the web browser for display. Data items are extracted from a header accompanying the web browser request. These include browser type and version, operating system, referring URL and IP address. These data items are supplemented with data items derived from the data items present in the header, such as location, time of day, day of week, and industry cycles.

Claims

1. A method for displaying targeted content on web pages by predicting the group membership of individual visitors, comprising the steps of:

(a) obtaining data items relating to a web browser request for a web page;

(b) applying the data items to a set of predictive transfer functions to obtain a series of confidence factors relating to a plurality of predefined visitor groups;

(c) selecting one of the plurality of predefined visitor groups based on the series of confidence factors;

(d) selecting content targeted to the selected group; and

(e) delivering the selected targeted content to the web browser for display.

2. The method of claim 1 wherein step (a) includes extracting visitor supplied data items from a header accompanying the web browser request.

3. The method of claim 2 wherein step (a) includes obtaining supplemental data items from lookup tables and/or databases.

4. The method of claim 3 wherein the step of obtaining supplemental data items includes obtaining location specific data items relating to an IP address of the visitor extracted from the header.

5. The method of claim 1 wherein the selecting steps (c) and (d) are performed by a best-fit matching algorithm.

6. The method of claim 5 wherein the best-fit matching algorithm is a least-squares fit algorithm.

7. The method of claim 1 wherein the predictive transfer functions are of the form:

Y_n=F(X₁ . . . X_m), where n is 1 to m predefined visitor groups, each Y is a confidence factor for membership in group n, and each X represents a different data item.

8. The method of claim 7 wherein the predictive transfer functions are of the form:

Y_n=aX₁+bX₂+cX₃ . . . , where n is 1 to m predefined visitor groups, each Y is a confidence factor for membership in group n, each X represents a different data item, and each a, b, c . . . , represents a weight factor for each X based on the relevance of each X to each Y_n.

9. The method of claim 1 further comprising applying statistical regression techniques to develop the predictive transfer functions to link anonymous characteristics of visitors to the plurality of predefined visitor groups, whose members have similar browsing patterns, and tagging and weighting content according to its relevance to each group.

10. The method of claim 1 wherein the targeted content is displayed on the first web page presented to the visitor.

11. The method of claim 2 wherein the visitor supplied data items include one or more of:

browser type, browser version, operating system, referring URL, and IP address.

12. The method of claim 3 wherein the supplemental data items include one or more of:

time of day, day of week, day of month, day of quarter, industry cycles, financial event schedule, and product event schedule.

13. The method of claim 4 wherein the location specific data items include one or more of:

country, time zone, city, state/province, top-level domain, secondary domain, line speed, and connection type.

14. An article of manufacture comprising:

a computer usable medium having computer readable program code means embodied therein for displaying targeted content on web pages by predicting the group membership of individual visitors, the computer readable program code means comprising;

computer readable program code means for obtaining data items relating to a web browser request for a web page;

computer readable program code means for applying the data items to a set of predictive transfer functions to obtain a series of confidence factors relating to a plurality of predefined visitor groups;

computer readable program code means for selecting one of the plurality of predefined visitor groups based on the series of confidence factors; and

computer readable program code means for selecting content targeted to the selected group; and

computer readable program code means for delivering the selected targeted content to the web browser for display.

15. The article of manufacture of claim 14 wherein the computer readable program code means for obtaining data items includes extracting visitor supplied data items from a header accompanying the web browser request.

16. The article of manufacture of claim 15 wherein the computer readable program code means for obtaining data items includes obtaining supplemental data items obtained from lookup tables and/or databases.

17. The article of manufacture of claim 16 wherein the computer readable program code means for obtaining supplemental data items includes obtaining location specific data items relating to an IP address of the visitor extracted from the header.

18. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform method steps for displaying targeted content on web pages by predicting the group membership of individual visitors, said method steps comprising:

(a) obtaining data items relating to a web browser request for a web page;

(b) applying the data items to a set of predictive transfer functions to obtain a series of confidence factors relating to a plurality of predefined visitor groups;

(c) selecting one of the plurality of predefined visitor groups based on the series of confidence factors;

(d) selecting content targeted to the selected group; and

(e) delivering the selected targeted content to the web browser for display.

19. The program storage device of claim 18 wherein step (a) includes extracting visitor supplied data items from a header accompanying the web browser request.

20. The program storage device of claim 19 wherein step (a) includes obtaining supplemental data items from lookup tables and/or databases.

21. The program storage device of claim 20 wherein the step of obtaining supplemental data items includes obtaining location specific data items relating to an IP address of the visitor extracted from the header.

22. An apparatus for displaying targeted content on web pages by predicting the group membership of individual visitors, comprising:

a network server with a memory; and

a computer program, performed by the network server, for;

(a) obtaining data items relating to a web browser request for a web page;

(b) applying the data items to a set of predictive transfer functions to obtain a series of confidence factors relating to a plurality of predefined visitor groups;

(c) selecting one of the plurality of predefined visitor groups based on the series of confidence factors;

(d) selecting content targeted to the selected group; and

(e) delivering the selected targeted content to the web browser for display.

23. The apparatus of claim 22 wherein step (a) includes extracting visitor supplied data items from a header accompanying the web browser request.

24. The apparatus of claim 23 wherein step (a) includes obtaining supplemental data items from lookup tables and/or databases.

25. The apparatus of claim 24 wherein the step of obtaining supplemental data items includes obtaining location specific data items relating to an IP address of the visitor extracted from the header.

26. The apparatus of claim 22 wherein the server includes a rules engine for storing the transfer functions and applying the data items to the transfer functions.

27. The apparatus of claim 26 wherein the rules engine stores a best-fit matching algorithm for performing the selecting steps (c) and (d).

28. The apparatus of claim 27 further including a content memory for storing a plurality of content and target group weights for the content.

29. The apparatus of claim 28 wherein the best fit algorithm selects the targeted content based on the target group weights obtained from the content memory and the confidence factors determined by the transfer functions.

30. The apparatus of claim 29 further comprising lookup tables containing the supplemental data items and wherein the rules engine queries the lookup tables for the supplemental data items.

31. The apparatus of claim 30 further comprising a geolocation database containing the location specific data items and wherein the network server performs a lookup based on the IP address of the visitor to obtain the location data specific to the visitor.

32. A method for predicting the group membership of individual web visitors for displaying targeted web content on web pages, comprising the steps of:

defining web visitor groups with similar browsing characteristics;

tagging each piece of a set of web content with a series of weights to indicate an applicability of each piece of content to each group;

defining data items relating to a web browser request for a web page;

identifying representative web content to be used to determine the group membership of a visitor;

defining a set of transfer functions linking the data items to the visitor groups, the transfer functions producing confidence factors that a particular visitor is considered a member of a particular group;

obtaining data items relating to a web browser request for a web page;

applying the data items to the set of transfer functions to obtain a series of confidence factors Yn relating to a plurality of predefined visitor groups;

selecting one of the plurality of predefined visitor groups based on the series of confidence factors;

selecting content targeted to the selected group; and

delivering the selected targeted content to the web browser for display.

33. The method of claim 32 wherein the data items include visitor supplied data items present in a header accompanying a web browser request.

34. The method of claim 33 wherein the data items include supplemental data items from lookup tables and/or databases.

35. The method of claim 34 wherein the supplemental data items include location specific data items relating to an IP address of the visitor.

36. The method of claim 35 wherein the selecting steps are performed by a best-fit matching algorithm.

37. The method of claim 36 wherein the visitor supplied data items include one or more of:

browser type, browser version, operating system, referring URL, and IP address.

38. The method of claim 37 wherein the supplemental data items include one or more of:

time of day, day of week, day of month, day of quarter, industry cycles, financial event schedule, and product event schedule.

39. The method of claim 38 wherein the location specific data items include one or more of:

country, time zone, city, state/province, top-level domain, secondary domain, line speed, and connection type.

40. The method of claim 32 wherein the set of transfer functions are of the form:

Y_n=F(X₁ . . . X_m), where n is 1 to m predefined visitor groups, each Y is a confidence factor for membership in group n, and each X represents a different data item.

41. The method of claim 32 wherein the set of transfer functions are of the form:

Y_n=aX₁+bX₂+cX₃ . . . , where n is 1 to m predefined visitor groups, each Y is a confidence factor for membership in group n, each X represents a different data item, and each a, b, c . . . , represents a weight factor for each X based on the relevance of each X to each Y_n.

42. The method of claim 32 comprising applying a statistical regression technique to define the set of transfer functions.

43. The method of claim 32, comprising applying a best guess estimation technique to define the set of transfer functions.

44. The method of claim 32, comprising applying a graphical analysis technique to define the set of transfer functions.

45. The method of claim 32, comprising applying a hypothesis testing technique to define the set of transfer functions.