MACHINE LEARNING BASED WEBINTERFACE PRODUCTION AND DEPLOYMENT SYSTEM

• US 20170192638A1
• Filed: 01/05/2017
• Published: 07/06/2017
• Est. Priority Date: 01/05/2016
• Status: Active Grant

• Current Assignee: Evolv Technology Solutions, Inc.
• Original Assignee: Sentient Technologies (Barbados) Limited
• Inventors: ISCOE, Neil, MIIKKULAINEN, Risto
• Granted Patent: US 11,386,318 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 11/3664 Environments for testing or debugging software
• CPC Class Codes: G06F 16/2379 Updates performed during online database operations; commit processing
• CPC Class Codes: G06F 16/26 Visual data mining; Browsing structured data
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
• CPC Class Codes: G06F 3/0484 for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
• CPC Class Codes: G06F 40/143 Markup, e.g. Standard Generalized Markup Language [SGML] or Document Type Definition [DTD]
• CPC Class Codes: G06F 8/36 Software reuse
• CPC Class Codes: G06F 9/451 Execution arrangements for user interfaces
• CPC Class Codes: G06N 3/04 Architecture, e.g. interconnection topology
• CPC Class Codes: G06N 3/06 Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
• CPC Class Codes: G06N 3/086 using evolutionary algorithms, e.g. genetic algorithms or genetic programming
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Roughly described, the technology disclosed provides a so-called machine learned conversion optimization (MLCO) system that uses evolutionary computations to efficiently identify most successful webpage designs in a search space without testing all possible webpage designs in the search space. The search space is defined based on webpage designs provided by marketers. Website funnels with a single webpage or multiple webpages are represented as genomes. Genomes identify different dimensions and dimension values of the funnels. The genomes are subjected to evolutionary operations like initialization, testing, competition, and procreation to identify parent genomes that perform well and offspring genomes that are likely to perform well. Each webpage is tested only to the extent that it is possible to decide whether it is promising, i.e., whether it should serve as a parent for the next generation, or should be discarded.

Claims

1. A computer-implemented webinterface production and deployment system, comprising:

a memory storing a candidate database having a population of candidate individuals, each of the candidate individuals identifying a funnel of one or more webinterfaces, each funnel having a plurality of dimensions along which a value can be specified, each candidate individual further identifying a value for each of the dimensions;

a presentation database which identifies a frontend element value for each of at least one available value of each of the dimensions of the funnel; and

a presentation module which applies a selected one of the candidate individuals to the presentation database to determine frontend element values corresponding to dimension values identified by the selected candidate individual, and which presents toward a user a funnel having the determined frontend element values.

2. The system of claim 1, wherein one of the available values for each of the dimensions is a default value,

wherein the funnel has a default frontend element value for the default dimension value of each of the dimensions, and

wherein the presentation module applies the selected candidate individual to the presentation database to determine frontend element values corresponding to all dimension values identified by the selected candidate individual other than the default dimension value for each of the dimensions.

3. A computer-implemented deployment system for a customizable webinterface funnel, the funnel being customizable along a plurality of dimensions, comprising:

a presentation database which identifies a frontend element value for each of at least one available value of each of the dimensions of the funnel;

a funnel customization specification memory storing, for a particular variation of the webinterface funnel, a value for each of the dimensions of the funnel; and

a presentation module which presents toward a user a funnel having frontend element values indicated by the presentation database as corresponding to the dimension values stored in the funnel customization specification memory.

4. The system of claim 3, wherein one of the available values for each of the dimensions is a default value,

wherein the funnel has a default frontend element value for the default dimension value of each of the dimensions, and

wherein the funnel presented toward the user by the presentation module has frontend element values indicated by the presentation database as corresponding to all of the dimension values identified in the funnel customization specification memory other than the default dimension value for each of the dimensions.

5. The system of claim 3, further configured to update the funnel customization specification memory with a new set of values for each of the dimensions of the funnel.

6. The system of claim 3, wherein the system retrieves the values for the funnel customization specification memory in response to user behavior for entering the funnel.

7. The system of claim 6, wherein the system also retrieves the presentation database in response to user behavior.

8. A computer-implemented webinterface production and deployment system, comprising:

a memory storing a database having a population of candidate individuals, each of the candidate individuals identifying a funnel of one or more webinterfaces, each funnel having a plurality of dimensions along which a value can be specified, each candidate individual further identifying a value for each of the dimensions; and

network server infrastructure which, in response to one or more first requests received over a network, returns toward a first client computer system;

a presentation database which identifies a frontend element value for each of at least one available value of each of the dimensions of the funnel,

a set of dimension values identified by a first selected one of the candidate individuals in the population, and

a presentation module which, in dependence upon the received set of dimension values, presents toward a user a funnel having frontend element values indicated by the received set of dimension values.

9. The system of claim 8, wherein one of the available values for each of the dimensions is a default value,

wherein the funnel has a default frontend element value for the default dimension value of each of the dimensions, and

wherein the presentation module applies the first selected candidate individual to the presentation database to determine frontend element values corresponding to all dimension values identified by the first selected candidate individual other than the default dimension value for each of the dimensions.

10. The system of claim 8, wherein the network server infrastructure, in response to one or more second requests received over a network, returns toward a second client computer system:

the presentation database;

the presentation module; and

the set of dimension values identified by a second selected one of the candidate individuals in the population.

11. The system of claim 8, wherein a webinterface represents a webpage.

12. The system of claim 8, wherein a webinterface represents a mobile application.

13. The system of claim 8, wherein a webinterface represents an e-mail.

14. The system of claim 8, wherein a webinterface represents a social media message.

15. The system of claim 8, wherein a webinterface represents a social media feed item.

16. The system of claim 8, wherein a webinterface represents a chat message.

17. The system of claim 8, wherein the population of candidate individuals in the database comprises candidate individuals being tested for their success in promoting predetermined target user behavior, the system further comprising:

a population initialization module which writes a preliminary pool of candidate individuals into the candidate individual population;

a candidate testing module that tests each of the candidate individuals in the candidate individual population by live online evaluation, the candidate testing module developing a performance measure for each of the candidate individuals undergoing live online evaluation, the performance measure being an indicator of success of the funnel in achieving the target user behavior;

a procreation module which adds to the candidate individual population new candidate individuals formed in dependence upon a respective set of one or more parent candidate individuals from the candidate individual population and the parents'"'"' performance measures; and

a control module that iterates the candidate testing module and the procreation module,

wherein the candidate testing module is included in the network server infrastructure and selects from the candidate individual population the selected candidate individuals whose dimension values are to be returned toward client computer systems in response to the requests.

18. The system of claim 17, wherein the target user behavior is conversion by users.

19. The system of claim 17, wherein the performance measure developed by the candidate testing module is based on conversion rates of the candidate individuals undergoing the live online evaluation.

20. The system of claim 17, wherein the performance measure developed by the candidate testing module is based on revenue received from the candidate individuals undergoing the live online evaluation.

21. The system of claim 8, wherein the network server infrastructure comprises:

a content delivery network (CDN) which, in response to a request from the client computer system, returns the presentation database and the presentation module toward the client computer system; and

a conversion system which retrieves the first selected candidate individual from the database and, in response to a request from the client computer system, returns toward the client computer system the set of dimension values identified by the first selected candidate individual.

22. A computer-implemented webinterface production and deployment system, comprising:

a memory storing a database having a population of candidate individuals, each of the candidate individuals identifying a funnel of one or more webinterfaces, each funnel having a plurality of dimensions along which a value can be specified, each candidate individual further identifying a value for each of the dimensions; and

a machine learned content delivery network having accessibly thereto a presentation database which identifies a frontend element value for each of at least one available value of each of the dimensions of the funnel, the machine learned content delivery network being configured to, in response to a request from a client computer system;

retrieve from the database a selected one of the candidate individuals in the population, and

delivers toward the client computer system a funnel having frontend element values indicated by the dimension values of the selected candidate individual.

23. The system of claim 22, wherein the machine learned content delivery network prepares the funnel in response to the request from a client computer system.

24. The system of claim 22, further comprising a conversion system having access to the database,

wherein the machine learned content delivery network, in retrieving the selected candidate individual, is configured further to request a candidate individual from the conversion system, and

wherein the conversion system is configured to select the selected candidate individual in response to a request from the machine learned content delivery network, and to return the selected candidate individual toward the machine learned content delivery network.

25. The system of claim 24, wherein the population of candidate individuals in the database comprises candidate individuals being tested for their success in promoting predetermined target user behavior, and wherein the conversion system comprises:

a population initialization module which writes a preliminary pool of candidate individuals into the candidate individual population;

a candidate testing module that tests each of the candidate individuals in the candidate individual population by live online evaluation, the candidate testing module developing a performance measure for each of the candidate individuals undergoing live online evaluation, the performance measure being an indicator of success of the funnel in achieving the target user behavior;

a procreation module which adds to the candidate individual population new candidate individuals formed in dependence upon a respective set of one or more parent candidate individuals from the candidate individual population and the parents'"'"' performance measures; and

a control module that iterates the candidate testing module and the procreation module,

wherein in selecting the selected candidate individual, the candidate testing module selects the selected candidate individual from the candidate individual population.

26. A computer-implemented webinterface production and deployment method, comprising:

providing a memory storing a candidate database having a population of a plurality of candidate individuals, each of the candidate individuals identifying a funnel of one or more webinterfaces, each funnel having a plurality of dimensions along which a value can be specified, each candidate individual further identifying a value for each of the dimensions;

providing a presentation database which identifies a frontend element value for each of at least one available value of each of the dimensions of the funnel;

applying a selected one of the candidate individuals to the presentation database to determine frontend element values corresponding to dimension values identified by the selected candidate individual; and

presenting toward a user a funnel having the determined frontend element values.