Method and system for creating a predictive model for targeting webpage to a surfer

• US 9,336,487 B2
• Filed: 06/24/2014
• Issued: 05/10/2016
• Est. Priority Date: 07/25/2008
• Status: Active Grant

• Current Assignee: LivePerson Incorporated
• Original Assignee: LivePerson Incorporated
• Inventors: Lahav, Shlomo
• Primary Examiner(s): Wong, Lut
• Application Number:
  US14/313,511
• Publication Number:
  US 20140310229A1
• Time in Patent Office: 686 Days
• Field of Search: 706/45
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9577 Optimising the visualization of content, e.g. distillation of HTML documents
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0251 Targeted advertisements

Abstract

System, methods, and computer-program products include receiving requests for a web page, retrieving predictive information related to the requests, and determining one or more predictive factors for an object presented with the web page, the one or more predictive factors being determined using the retrieved predictive information. The systems, methods, and computer-program products further include generating a plurality of predictive models for the object using the one or more predictive factors, determining a score for each predictive model, selecting a group of predictive models from the plurality of predictive models using the score of each predictive model in the group, and generating a representative predictive model for the object using the group of predictive models, the representative predictive model being associated with the object.

Claims

1. A computer-implemented method, comprising:

receiving, by a computing device, a plurality of requests for a web page from a plurality of sources, wherein the web page includes a configuration defining one or more objects presented with the web page, wherein each object is associated with a predictive model including one or more predictive factors, wherein the one or more predictive factors are determined using the plurality of received requests, wherein the one or more predictive factors are used to calculate a predictive value, and wherein a predictive value corresponds to a likelihood that an object will be selected by any one of the plurality of sources when presented in the configuration of the web page;

retrieving predictive information related to the plurality of requests;

determining one or more predictive factors for an object presented with the web page, wherein the one or more predictive factors are determined using the retrieved predictive information;

generating a plurality of predictive models for the object using the one or more predictive factors;

determining a score for each predictive model;

selecting a group of predictive models from the plurality of predictive models, wherein the group of predictive models are selected using the score of each predictive model in the group; and

generating a representative predictive model for the object using the group of predictive models, wherein the representative predictive model is associated with the object.

2. The method of claim 1, wherein generating the representative predictive model for the object using the group of predictive models includes calculating an average value of coefficients associated with the group of predictive models.

3. The method of claim 1, further comprising:

receiving a subsequent request for the web page from a source, wherein the subsequent request is received after the plurality of requests, and wherein the subsequent request includes subsequent predictive information;

retrieving the representative predictive model associated with the object;

retrieving one or more predictive factors for the object, wherein the one or more predictive factors are associated with the subsequent predictive information;

determining a predictive value for the object using the representative predictive model, wherein the predictive value is determined by inputting the one or more retrieved predictive factors; and

selecting the object for inclusion in the configuration of the web page, wherein the object is selected using the predictive value.

4. The method of claim 1, wherein generating the plurality of predictive models includes performing logistic regression, linear regression, or decision tree analysis.

5. The method of claim 1, wherein the predictive information includes behavioral information and information associated with the request for the web page.

6. The method of claim 5, wherein the behavioral information includes a time of one or more previous visits in a domain that includes the requested web page, a time that the web page was last requested, a number of visits in the domain, a number of times that the object has been presented, or a number of times that the object has been selected.

7. The method of claim 6, wherein the behavioral information is stored in a cookie.

8. The method of claim 5, wherein the information associated with the request for the web page includes a receipt time of the request, a uniform resource locator key associated with the request, an internet protocol address, or a type of browser application.

9. A system, comprising:

a processor; and

a non-transitory computer-readable storage medium containing instructions configured to cause the processor to perform operations including;

receiving a plurality of requests for a web page from a plurality of sources, wherein the web page includes a configuration defining one or more objects presented with the web page, wherein each object is associated with a predictive model including one or more predictive factors, wherein the one or more predictive factors are determined using the plurality of received requests, wherein the one or more predictive factors are used to calculate a predictive value, and wherein a predictive value corresponds to a likelihood that an object will be selected by any one of the plurality of sources when presented in the configuration of the web page;

retrieving predictive information related to the plurality of requests;

determining one or more predictive factors for an object presented with the web page, wherein the one or more predictive factors are determined using the retrieved predictive information;

generating a plurality of predictive models for the object using the one or more predictive factors;

determining a score for each predictive model;

selecting a group of predictive models from the plurality of predictive models, wherein the group of predictive models are selected using the score of each predictive model in the group; and

generating a representative predictive model for the object using the group of predictive models, wherein the representative predictive model is associated with the object.

10. The system of claim 9, wherein generating the representative predictive model for the object using the group of predictive models includes calculating an average value of coefficients associated with the group of predictive models.

11. The system of claim 9, further comprising instructions configured to cause the processor to perform operations including:

receiving a subsequent request for the web page from a source, wherein the subsequent request is received after the plurality of requests, and wherein the subsequent request includes subsequent predictive information;

retrieving the representative predictive model associated with the object;

retrieving one or more predictive factors for the object, wherein the one or more predictive factors are associated with the subsequent predictive information;

determining a predictive value for the object using the representative predictive model, wherein the predictive value is determined by inputting the one or more retrieved predictive factors; and

selecting the object for inclusion in the configuration of the web page, wherein the object is selected using the predictive value.

12. The system of claim 9, wherein generating the plurality of predictive models includes performing logistic regression, linear regression, or decision tree analysis.

13. The system of claim 9, wherein the predictive information includes behavioral information and information associated with the request for the web page.

14. The system of claim 13, wherein the behavioral information includes a time of one or more previous visits in a domain that includes the requested web page, a time that the web page was last requested, a number of visits in the domain, a number of times that the object has been presented, or a number of times that the object has been selected.

15. The system of claim 14, wherein the behavioral information is stored in a cookie.

16. The system of claim 13, wherein the information associated with the request for the web page includes a receipt time of the request, a uniform resource locator key associated with the request, an internet protocol address, or a type of browser application.

17. A computer-program product, tangibly embodied in a non-transitory machine-readable medium, including instructions configured to cause a data processing apparatus to:

receive a plurality of requests for a web page from a plurality of sources, wherein the web page includes a configuration defining one or more objects presented with the web page, wherein each object is associated with a predictive model including one or more predictive factors, wherein the one or more predictive factors are determined using the plurality of received requests, wherein the one or more predictive factors are used to calculate a predictive value, and wherein a predictive value corresponds to a likelihood that an object will be selected by any one of the plurality of sources when presented in the configuration of the web page;

retrieve predictive information related to the plurality of requests;

determine one or more predictive factors for an object presented with the web page, wherein the one or more predictive factors are determined using the retrieved predictive information;

generate a plurality of predictive models for the object using the one or more predictive factors;

determine a score for each predictive model;

select a group of predictive models from the plurality of predictive models, wherein the group of predictive models are selected using the score of each predictive model in the group; and

generate a representative predictive model for the object using the group of predictive models, wherein the representative predictive model is associated with the object.

18. The computer-program product of claim 17, wherein generating the representative predictive model for the object using the group of predictive models includes calculating an average value of coefficients associated with the group of predictive models.

19. The computer-program product of claim 17, further comprising instructions configured to cause the data processing apparatus to:

receive a subsequent request for the web page from a source, wherein the subsequent request is received after the plurality of requests, and wherein the subsequent request includes subsequent predictive information;

retrieve the representative predictive model associated with the object;

retrieve one or more predictive factors for the object, wherein the one or more predictive factors are associated with the subsequent predictive information;

determine a predictive value for the object using the representative predictive model, wherein the predictive value is determined by inputting the one or more retrieved predictive factors; and

select the object for inclusion in the configuration of the web page, wherein the object is selected using the predictive value.

20. The computer-program product of claim 17, wherein generating the plurality of predictive models includes performing logistic regression, linear regression, or decision tree analysis.