SYSTEM AND METHOD FOR COLLECTING AND TARGETING VISITOR BEHAVIOR

• US 20090248494A1
• Filed: 03/31/2009
• Published: 10/01/2009
• Est. Priority Date: 04/01/2008
• Status: Active Grant

• Current Assignee: Certona Corporation (StartApp, Inc.)
• Original Assignee: Certona Corporation (StartApp, Inc.)
• Inventors: Sheik, Meyar, Hueter, Geoffrey J., Hueter, Noble H., Quandt, Steven C.
• Granted Patent: US 9,916,611 B2
• Time in Patent Office: Days
• US Class Current: 705/10
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/0185 Product, service or business identity fraud
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0204 Market segmentation
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: H04L 51/212 using filtering or selective blocking

Abstract

A system and method is disclosed for collecting website visitor activity for profiling visitor interests and dynamically modifying the content of the website to better match the visitor'"'"'s profile. The visitor activity data is collected directly from the visitor'"'"'s client browser or from the website'"'"'s own web log information. The collected data consists of the page identifier, page links, and the previous page identifier. Similarly, the modified page content can be sent directly to the client browser or can be sent back to the website server for integration with the other page content. The collected data is stored in a database. Based on the amount of information collected on the visitor and the various items that are presented on the website, the visitors and items are profiled so that a visitor'"'"'s response to other items can be predicted and recommended to the visitor. The recommendations can be requested and displayed directly by and to the client browser or the website server can make the request and subsequently display the matching content. The system has application in personalization, behavioral targeting, Internet retailing, social networking, affiliate marketing, and online advertising, to name but a few applications.

Claims

1. A system for collecting and targeting visitor behavior, comprising:

(a) an event target model for relating subject behavior to the intended outcome of the application;

(b) a service system to collect subject behavior and provide recommendations;

(c) a test system for assigning subjects to different test groups and measuring their response to recommendations;

(d) subjects and objects represented in vector form;

(e) object vectors generated from object-object coincidences;

(f) subject vectors generated from subject-object ratings or affinities;

(g) recommendations to, and collecting data from an external application; and

(h) a cross-referencing engine for subjects and objects, to allow applications to manage subjects and objects, while still obtaining the aggregated benefit of a common database

whereby said system for collecting and targeting visitor behavior is both session based and ratings based, and allows deterministic estimate of subject-object affinities from behavioral data.

2. The system for collecting and targeting visitor behavior, according to claim 1, wherein said event target model for relating subject behavior to the intended outcome of the application includes a target model derived from the probabilities or historical frequencies of a subject-object event leading to the ultimate achievement of the application'"'"'s target event for that subject and object, with events weighted by the profiling engine according to the probability of an event leading to a target event.

3. The system for collecting and targeting visitor behavior, according to claim 1, wherein said service system to collect subject behavior and provide recommendations includes a system for collecting data employing application scripting, which is embedded into the application which is deploying the recommendations, or the use of web services to generate recommendations.

4. The system for collecting and targeting visitor behavior, according to claim 3, wherein said application scripting includes application scripting, which creates user subject identities, randomly assigns them to test groups, including on a sampling basis, extracts subject object events via pattern matching to application data, and displays system-generated recommendations to the appropriate test group, either through web services are directly to a browser by employing content overwrite to generate dynamic content on web pages, including cached web pages.

5. The system for collecting and targeting visitor behavior, according to claim 4, wherein said application scripting includes Javascript which is invoked by the HTML which renders a web page on a web browser and which stores the subject user identifier in a cookie.

6. The system for collecting and targeting visitor behavior, according to claim 5, wherein said subject user identifier cookie includes a first party cookie or a third party cookie, and further wherein the visitor profile storage is distributed by storing the subject profile in said subject user identifier cookie.

7. The system for collecting and targeting visitor behavior, according to claim 1, wherein said object vectors include object vectors generated solely from object-object coincidences whereby subject object ratings are not required, and further wherein profiles are generated without negative ratings from association-only data, employing seasonal memory through variable updating, seasonal memory and differentiating cross sells and up sells through the temporal asymmetry of antecedent consequent associations.

8. The system for collecting and targeting visitor behavior, according to claim 1, wherein said subject vectors include subject vectors generated by combining the object vectors with inferred rating levels for different levels of the target model and further wherein said inferred rating levels are estimated by the rating levels predicted by matching subject and object vectors.

9. The system for collecting and targeting visitor behavior, according to claim 1, wherein said objects includes objects profiled across types of objects and subjects and across time and category, which may be cross-associated from multiple applications or catalogs based on subjects interacting with multiple applications and catalogs, and the catalogs containing overlapping objects between applications.

10. The system for collecting and targeting visitor behavior, according to claim 9, wherein said catalogs includes catalog containing disparate types of objects.

11. A method for collecting and targeting visitor behavior, comprising the steps of:

(a) providing an event target model for relating subject behavior to the intended outcome of the application;

(b) providing a service system to collect subject behavior and provide recommendations;

(c) providing a test system for assigning subjects to different test groups and measuring their response to recommendations;

(d) representing subjects and objects in vector form;

(e) generating object vectors from object-object coincidences;

(f) generating subject vectors from subject-object ratings or affinities;

(g) sending recommendations to, and collecting data from an external application; and

(h) providing a cross-referencing engine for subjects and objects, to allow applications to manage subjects and objects, while still obtaining the aggregated benefit of a common database

whereby said system for collecting and targeting visitor behavior is both session based and ratings based, and allows deterministic estimate of subject-object affinities from behavioral data.

12. The method for collecting and targeting visitor behavior, according to claim 11, wherein said step of providing an event target model for relating subject behavior to the intended outcome of the application includes the step of providing a target model wherein said event target model for relating subject behavior to the intended outcome of the application includes a target model derived from the probabilities or historical frequencies of a subject-object event leading to the ultimate achievement of the application'"'"'s target event for that subject and object, with events weighted by the profiling engine according to the probability of an event leading to a target event.

13. The method for collecting and targeting visitor behavior, according to claim 11, wherein said step of providing a service system to collect subject behavior and provide recommendations includes the step of providing a service system wherein said service system to collect subject behavior and provide recommendations includes a system for collecting data employing application scripting, which is embedded into the application which is deploying the recommendations, or the use of web services to generate recommendations.

14. The method for collecting and targeting visitor behavior, according to claim 13, wherein said step of providing a service system to collect subject behavior and provide recommendations includes providing a service system wherein said application scripting includes application scripting, which creates user subject identities, randomly assigns them to test groups, including on a sampling basis, extracts subject object events via pattern matching to application data, and displays system-generated recommendations to the appropriate test group, either through web services are directly to a browser by employing content overwrite to generate dynamic content on web pages, including cached web pages.

15. The method for collecting and targeting visitor behavior, according to claim 14, wherein said step of providing a service system includes the step of providing a service system wherein said application scripting includes Javascript which is invoked by the HTML which renders a web page on a web browser and which stores the subject user identifier in a cookie.

16. The method for collecting and targeting visitor behavior, according to claim 15, wherein said step of generating object vectors from object-object coincidences includes the step of generating object vectors from object-object coincidences wherein said subject user identifier cookie includes a first party cookie or a third party cookie, and further wherein the visitor profile storage is distributed by storing the subject profile in said subject user identifier cookie.

17. The method for collecting and targeting visitor behavior, according to claim 11, wherein said step of generating subject vectors from subject-object ratings or affinities includes the step of generating subject vectors from subject-object ratings or affinities wherein said object vectors include object vectors generated solely from object-object coincidences whereby subject object ratings are not required, and further wherein profiles are generated without negative ratings from association-only data, employing seasonal memory through variable updating, seasonal memory and differentiating cross sells and up sells through the temporal asymmetry of antecedent consequent associations.

18. The method for collecting and targeting visitor behavior, according to claim 11, wherein said step of generating object vectors includes the step of generating object vectors wherein said subject vectors include subject vectors generated by combining the object vectors with inferred rating levels for different levels of the target model and further wherein said inferred rating levels are estimated by the rating levels predicted by matching subject and object vectors.

19. A method for using a system for collecting and targeting visitor behavior, comprising the steps of:

(a) creating a target model;

(b) collecting user subject object behavior data using an application script;

(c) storing user subject object behavior data;

(d) generating subject and object vectors from said stored subject object behavior data;

(e) storing said vectors; and

(f) scoring said stored object vectors against said stored subject and object vectors to generate recommendations.

20. The method for using a system for collecting and targeting visitor behavior, according to claim 19, wherein said step of scoring said stored object vectors against said stored subject and vectors to generate recommendations includes scoring said stored object vectors against said stored subject and vectors to generate recommendations within a personalized shopping website application, including personalized product placements, advertisements, reviews, videos, social networking sites, personalized data feeds, RSS, referring websites, ad networks and other content.