IDENTIFICATION OF ENTITIES LIKELY TO ENGAGE IN A BEHAVIOR

• US 20130110746A1
• Filed: 11/01/2011
• Published: 05/02/2013
• Est. Priority Date: 11/01/2011
• Status: Active Grant

• Current Assignee: Accenture Global Services Limited (Accenture PLC)
• Original Assignee: Accenture Global Services Limited (Accenture PLC)
• Inventors: AHN, Junghoon
• Granted Patent: US 8,843,421 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: H04L 43/10 Active monitoring, e.g. heartbeat, ping or trace-route
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

To entities that are likely to engage in a behavior, a behavior discrimination threshold for each of a plurality of behavior variables is determined. Variable values for each entity of a plurality of target entities and for each of the behavior variables are compared to the relevant behavior discrimination threshold to provide identifying variables for each entity. An index value is determined for each of the plurality of target entities based on the identifying variables for that entity. An index threshold is determined such that those entities of the target entities having index values that meet or exceed the index threshold are likely to exhibit or engage in the behavior more frequently than a random sampling of the plurality of training entities. Actions designed to affect the likelihood that the resulting targeted entities will engage in the behavior may be directed to the targeted entities.

Claims

1. A method for identifying entities likely to engage in a behavior, the method comprising:

identifying a behavior discrimination threshold for each variable of a plurality of behavior variables;

for each entity of a plurality of target entities and for each variable of the plurality of behavior variables, determining, by a processing device, that a variable value of the entity satisfies the behavior discrimination threshold for the variable to thereby identify the variable as an identifying variable for the entity;

for each entity of the plurality of target entities, determining, by the processing device, an index value for the entity according to the identifying variables for the entity; and

determining, by the processing device, an index threshold wherein those entities of the plurality of the target entities having index values at least meeting the index threshold are likely to exhibit the behavior more frequently than a random sampling of a plurality of training entities.

2. The method of claim 1, wherein identifying the behavior discrimination threshold for each variable further comprises identifying, by the processing device, the behavior discrimination threshold based on training data for the plurality of training entities.

3. The method of claim 2, wherein identifying the behavior discrimination threshold for each variable further comprises:

determining, by the processing device based on the training data, a distribution for the variable based on that portion of the plurality of training entities exhibiting the behavior; and

selecting, by the processing device, as the behavior discriminating threshold a value of the distribution that delineates at least a portion of that portion of the plurality of training entities exhibiting the behavior.

4. The method of claim 3, wherein the value of the distribution that delineates at least a portion of those entities of the plurality of training entities exhibiting the behavior is determined according to a mean value of the distribution extended by a delta value.

5. The method of claim 4, wherein the delta value is one standard error of the distribution.

6. The method of claim 1, wherein determining the index threshold further comprises selecting the index threshold based on a candidate index value at which those entities of the plurality of target entities at least meeting the candidate index value exhibit the behavior more frequently by a desired factor as compared to the random sampling of the plurality of training entities.

7. The method of claim 1, wherein the plurality of training entities and the plurality of target entities are subscribers to a service and the behavior is discontinuing the service.

8. The method of claim 1, further comprising:

identifying, by the processing device, those entities of the plurality of target entities having an index value at least meeting the index threshold to provide targeted entities; and

taking at least one action relative to the targeted entities to affect the likelihood that the targeted entities will exhibit the behavior.

9. An apparatus for identifying entities likely to engage in a behavior, comprising:

a processor;

a storage device, operatively connected to the processing device and having stored thereon instructions that, when executed by the processor, cause the processor to;

identify a behavior discrimination threshold for each variable of a plurality of behavior variables;

for each entity of a plurality of target entities and for each variable of the plurality of behavior variables, determine that a variable value of the entity satisfies the behavior discrimination threshold for the variable to thereby identify the variable as an identifying variable for the entity;

for each entity of the plurality of target entities, determine an index value for the entity according to the identifying variables for the entity; and

determine an index threshold wherein those entities of the plurality of the target entities having index values at least meeting the index threshold are likely to exhibit the behavior more frequently than a random sampling of a plurality of training entities.

10. The apparatus of claim 9, wherein those instructions that, when executed by the processor, cause the processor to identify the behavior discrimination threshold for each variable are further operative to cause the processor to identify the behavior discrimination threshold based on training data for the plurality of training entities.

11. The apparatus of claim 10, wherein those instructions that, when executed by the processor, cause the processor to identify the behavior discrimination threshold for each variable are further operative to cause the processor to:

determine, based on the training data, a distribution for the variable based on that portion of the plurality of training entities exhibiting the behavior; and

select as the behavior discriminating threshold a value of the distribution that delineates at least a portion of that portion of the plurality of training entities exhibiting the behavior.

12. The apparatus of claim 11, wherein those instructions that, when executed by the processor, cause the processor to identify the behavior discrimination threshold for each variable are further operative to cause the processor to:

determine the value of the distribution that delineates at least a half of that portion of the plurality of training entities exhibiting the behavior according to a mean value of the distribution extended by a delta value.

13. The apparatus of claim 12, wherein those instructions that, when executed by the processor, cause the processor to identify the behavior discrimination threshold for each variable are further operative to cause the processor to:

calculate a standard error of the distribution, wherein the delta value is the standard error.

14. The apparatus of claim 9, wherein those instructions that, when executed by the processor, cause the processor to determine the index threshold are further operative to cause the processor to:

select the index threshold based on a candidate index value at which those entities of the plurality of target entities at least meeting the candidate index value exhibit the behavior more frequently by a desired factor as compared to the random sampling of the plurality of training entities.

15. The apparatus of claim 9, wherein the plurality of training entities and the plurality of target entities are subscribers to a service and the behavior is discontinuing the service.

16. The apparatus of claim 9, wherein the storage device further comprises instructions that, when executed by the processor, cause the processor to:

identify those entities of the plurality of target entities having an index value at least meeting the index threshold to provide targeted entities; and

provide the targeted entities as output.

17. An apparatus for identifying entities likely to engage in a behavior, comprising:

at least one data source comprising data for a plurality of training entities and for a plurality of target entities; and

a behavior identification processor, implemented by at least one processing device, operative to;

identify a behavior discrimination threshold for each variable of a plurality of behavior variables;

determine, for each entity of the plurality of target entities and for each variable of the plurality of behavior variables, that a variable value of the entity satisfies the behavior discrimination threshold for the variable to thereby identify the variable as an identifying variable for the entity;

determine, for each entity of the plurality of target entities, an index value for the entity according to the identifying variables for the entity; and

determine an index threshold wherein those entities of the plurality of the target entities having index values at least meeting the index threshold are likely to exhibit the behavior more frequently than a random sampling of a plurality of training entities.

18. The apparatus of claim 17, wherein the at least one data source comprises a plurality of data sources.