Method, apparatus, and computer program product for predictive dynamic bidding rule generation for digital content objects

1. An apparatus for predictive dynamic generation of a digital content object bidding rule, the apparatus comprising at least one processor and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to:

retrieve a plurality of device rendered object data structures from a plurality of data streams, each data stream retrieved from one of a plurality of data sources, the plurality of device rendered object data structures associated with a plurality of device rendered object transactions;

for each device rendered object data structure of the plurality of device rendered object data structures,

extract one or more digital content objects contained in the device rendered object data structure, one or more device rendered object attributes, one or more device rendered object interaction currency value, and one or more device rendered object interaction timestamp;

aggregate the one or more digital content objects on a per digital content object basis into an aggregated dataset based at least in part on aggregation of the one or more digital content objects associated with each of the plurality of device rendered object transactions;

append, for each associated digital content object in the aggregated dataset, one or more of the device rendered object attributes, device rendered object interaction currency values, and one or more device rendered object timestamps to a digital content object data structure associated with the digital content object;

define, using a machine learning model, a plurality of digital content object clusters, wherein each digital content object cluster is defined based upon similarities identified by the machine learning model;

upon determining that an average digital content object interaction currency value associated with a digital content object cluster exceeds an average digital content object interaction currency value associated with a digital content object superset, generate an electronic bid adjustment rule to be applied to each existing electronic bid value for each digital content object of the digital content object cluster; and

adjust the existing electronic bid value for each digital content object of the digital content object cluster based on the electronic bid adjustment rule.

2. The apparatus of claim 1 , wherein the plurality of data sources comprises one or more of a device rendered object attributes repository, a device rendered object transactions repository, and a digital content object interaction repository.

3. The apparatus of claim 1 , wherein the average digital content object interaction currency value is calculated based upon historical digital content object interaction currency values associated with each digital content object of the related digital content object cluster.

4. The apparatus of claim 1 , wherein a digital content object is a keyword.

5. The apparatus of claim 1 , wherein the plurality of digital content object clusters is continuously updated based on data streams received in real-time.

6. The apparatus of claim 1 , wherein the machine learning model is based on one of a classification algorithm, a clustering algorithm, or a decision tree.

7. The apparatus of claim 1 , wherein the electronic bid value is for transmitting to a third party content provider for use in an electronic digital content auction.

8. The apparatus of claim 1 , wherein the device rendered object data structure comprises one or more device rendered object attributes, one or more associated digital content objects, and one or more device rendered object interaction currency values.

9. The apparatus of claim 1 , wherein the digital content object data structure comprises one or more device rendered object attributes, one or more device rendered object interaction currency values, and one or more device rendered object interaction timestamps.

10. The apparatus of claim 1 , wherein the digital content objects superset comprises all digital content objects known to a device rendered object service.

11. The apparatus of claim 1 , wherein the electronic bid adjustment rule comprises one or more device rendered object attributes and one or more device rendered object interaction currency values.

12. The apparatus of claim 1 , wherein the one or more digital content objects are aggregated based at least in part on category data for each of the plurality of device rendered object data structures, location data for each of the plurality of device rendered object data structures, or a shared service attribute for each of the plurality of device rendered object data structures.

13. The apparatus of claim 1 , wherein the machine learning model is trained based at least in part on at least one clustered attribute.

14. The apparatus of claim 1 , wherein the machine learning model comprises a decision tree, wherein to generate the machine learning model the apparatus is caused to:

generate a node for a first attribute of the one or more device rendered object attributes; and

expanding the decision tree by adding at least one node corresponding to another attribute of the one or more device rendered object attributes until one or more clusters of digital content objects is associated with a calculated average digital content object interaction currency value that exceeds a threshold,

wherein the electronic bid adjustment rule is generated based at least in part on a set of parent nodes of the decision tree in response to determination that the calculated average digital content object interaction currency value exceeds the threshold.

15. The apparatus of claim 1 , wherein the average digital content object interaction currency value is determined based at least in part on a plurality of digital content objects, aggregated from various different device rendered object data structures, representing a related digital content object cluster.

16. A system comprising one or more processors and one or more non-transitory storage media for storing instructions that, when executed by the one or more processors, cause the system to:

retrieve a plurality of device rendered object data structures from a plurality of data streams, each data stream retrieved from one of a plurality of data sources, the plurality of device rendered object data structures associated with a plurality of device rendered object transactions;

for each device rendered object data structure of the plurality of device rendered object data structures,

extract one or more digital content objects contained in the device rendered object data structure, one or more device rendered object attributes, one or more device rendered object interaction currency value, and one or more device rendered object interaction timestamp;

aggregate the one or more digital content objects on a per digital content object basis into an aggregated dataset based at least in part on aggregation of the one or more digital content objects associated with each of the plurality of device rendered object transactions;

append, for each associated digital content object, in the aggregated dataset, one or more of the device rendered object attributes, device rendered object interaction currency values, and one or more device rendered object timestamps to a digital content object data structure associated with the digital content object;

define, using a machine learning model, a plurality of digital content object clusters, wherein each digital content object cluster is defined based upon similarities identified by the machine learning model;

upon determining that an average digital content object interaction currency value associated with a digital content object cluster exceeds an average digital content object interaction currency value associated with a digital content object superset, generate an electronic bid adjustment rule to be applied to each existing electronic bid value for each digital content object of the digital content object cluster; and

adjust the existing electronic bid value for each digital content object of the digital content object cluster based on the electronic bid adjustment rule.

17. The system of claim 16 , wherein the plurality of data sources comprises one or more of a device rendered object attributes repository, a device rendered object transactions repository, and a digital content object interaction repository.

18. The system of claim 16 , wherein the average digital content object interaction currency value is calculated based upon historical digital content object interaction currency values associated with each digital content object of the related digital content object cluster.

19. The system of claim 16 , wherein a digital content object is a keyword.

20. The system of claim 16 , wherein the plurality of digital content object clusters is continuously updated based on data streams received in real-time.

21. The system of claim 16 , wherein the machine learning model is based on one of a classification algorithm, a clustering algorithm, or a decision tree.

22. The system of claim 16 , wherein the electronic bid value is for transmitting to a third party content provider for use in an electronic digital content auction.

23. A computer-implemented method, comprising:

retrieving, by a processor, a plurality of device rendered object data structures from a plurality of data streams, each data stream retrieved from one of a plurality of data sources, the plurality of device rendered object data structures associated with a plurality of device rendered object transactions;

for each device rendered object data structure of the plurality of device rendered object data structures,

extracting, by the processor, one or more digital content objects contained in the device rendered object data structure, one or more device rendered object attributes, one or more device rendered object interaction currency value, and one or more device rendered object interaction timestamp;

aggregating the one or more digital content objects on a per digital content object basis into an aggregated dataset based at least in part on aggregation of the one or more digital content objects associated with each of the plurality of device rendered object transactions;

appending, for each associated digital content object in the augmented dataset, by the processor, one or more of the device rendered object attributes, device rendered object interaction currency values, and one or more device rendered object timestamps to a digital content object data structure associated with the digital content object;

defining, by the processor and using a machine learning model, a plurality of digital content object clusters, wherein each digital content object cluster is defined based upon similarities identified by the machine learning model;

upon determining that an average digital content object interaction currency value associated with a digital content object cluster exceeds an average digital content object interaction currency value associated with a digital content object superset, generating, by the processor, an electronic bid adjustment rule to be applied to each existing electronic bid value for each digital content object of the digital content object cluster; and

adjusting, by the processor, the existing electronic bid value for each digital content object of the digital content object cluster based on the electronic bid adjustment rule.