Method, apparatus, and computer program product for inferring device rendered object interaction behavior
The present disclosure relates to inferring device rendered object interaction behavior. An apparatus is configured to receive historical device rendered object interaction data associated with a client device. The historical device rendered object interaction data includes a plurality of records. Each record is associated with an device rendered object category. For each record, the device rendered object category is extracted. For each device rendered object category, a penalty value is generated. The penalty value is a programmatically generated predicted network time delay during which device rendered objects associated with the device rendered object category are not to be transmitted to the client device. A penalty decision, used to determine whether to apply the penalty value to future transmissions of device rendered objects to the client device, is generated.
1. An apparatus for inferring repeated device rendered object interaction behavior, 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:
receive historical device rendered object interaction data associated with a client device, the client device associated with a user having a user profile, the historical device rendered object interaction data comprising a plurality of records, wherein each record of the historical device rendered object interaction data is associated with a device rendered object category;
for each record of the historical device rendered object interaction data, extract the device rendered object category;
for each extracted device rendered object category,
generate a penalty value, wherein the penalty value is a programmatically generated predicted network time delay during which device rendered objects associated with the device rendered object category are not to be transmitted to the client device;
generate a penalty decision, wherein the penalty decision is used to determine whether to apply the penalty value to future transmissions of device rendered objects to the client device;
retrieve a plurality of device rendered objects selected for transmission to the client device;
rank the plurality of device rendered objects according to a ranking algorithm;
for each device rendered object of the plurality of device rendered objects,
in the event a penalty decision associated with the device rendered object category of the device rendered object indicates that its penalty value should be applied, generate a penalty adjusted ranking value by applying the penalty value to a ranking value of the device rendered object; and
re-rank the plurality of device rendered objects according to the penalty adjusted ranking value; and
transmit the re-ranked plurality of device rendered objects to the client device for rendering via a display of the client device, wherein the plurality of device rendered objects is arranged for display according to the penalty adjusted ranking values of each device rendered object.
2. The apparatus of claim 1 , wherein the penalty decision is generated using a logistic regression model.
3. The apparatus of claim 1 , wherein the penalty value is generated using a machine learning model.
4. The apparatus of claim 1 , wherein the ranking algorithm assigns a ranking value to each device rendered object of the plurality of device rendered objects based on one or more of users' preferences, the user's age group, the user's gender, the user's past purchases and clicks, popularity of objects, object attributes, object text title and descriptions, preference of the user for similar objects, distance between the object and the user.
5. The apparatus of claim 4 , wherein the ranking algorithm comprises a plurality of ranking stages.
6. A system for inferring device rendered object interaction behavior, the system comprising a server in communication with a network, the server 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 server to:
receive historical device rendered object interaction data associated with a client device, the client device associated with a user having a user profile, the historical device rendered object interaction data comprising a plurality of records, wherein each record of the historical device rendered object interaction data is associated with a device rendered object category;
for each record of the historical device rendered object interaction data, extract the device rendered object category;
for each extracted device rendered object category,
generate a penalty value, wherein the penalty value is a programmatically generated predicted network time delay during which device rendered objects associated with the device rendered object category are not to be transmitted to the client device;
generate a penalty decision, wherein the penalty decision is used to determine whether to apply the penalty value to future transmissions of device rendered objects to the client device;
retrieve a plurality of device rendered objects selected for transmission to the client device;
rank the plurality of device rendered objects according to a ranking algorithm;
for each device rendered object of the plurality of device rendered objects,
in the event a penalty decision associated with the device rendered object category of the device rendered object indicates that its penalty value should be applied, generate a penalty adjusted ranking value by applying the penalty value to a ranking value of the device rendered object; and
re-rank the plurality of device rendered objects according to the penalty adjusted ranking value; and
transmit the re-ranked plurality of device rendered objects to the client device for rendering via a display of the client device, wherein the plurality of device rendered objects is arranged for display according to the penalty adjusted ranking values of each device rendered object.
7. The system of claim 6 , wherein the penalty decision is generated using a logistic regression model.
8. The system of claim 6 , wherein the penalty value is generated using a machine learning model.
9. The system of claim 6 , wherein the ranking algorithm assigns a ranking value to each device rendered object of the plurality of device rendered objects based on one or more of users' preferences, the user's age group, the user's gender, the user's past purchases and clicks, popularity of objects, object attributes, object text title and descriptions, preference of the user for similar objects, distance between the object and the user.
10. The system of claim 9 , wherein the ranking algorithm comprises a plurality of ranking stages.
11. The system of claim 9 , wherein object attributes comprise price and inception date.
12. A computer-implemented method for inferring device rendered object interaction behavior, the method comprising:
receiving, using a processor, historical device rendered object interaction data associated with a client device, the client device associated with a user having a user profile, the historical device rendered object interaction data comprising a plurality of records, wherein each record of the historical device rendered object interaction data is associated with a device rendered object category;
for each record of the historical device rendered object interaction data, extract the device rendered object category;
for each device rendered object category,
generating, using the processor, a penalty value, wherein the penalty value is a programmatically generated predicted network time delay during which device rendered objects associated with the device rendered object category are not to be transmitted to the client device;
generating, using the processor, a penalty decision, wherein the penalty decision is used to determine whether to apply the penalty value to future transmissions of device rendered objects to the client device;
retrieving, using the processor, a plurality of device rendered objects selected for transmission to the client device;
ranking, using the processor, the plurality of device rendered objects according to a ranking algorithm;
for each device rendered object of the plurality of device rendered objects,
in the event a penalty decision associated with the device rendered object category of the device rendered object indicates that its penalty value should be applied, generate a penalty adjusted ranking value by applying the penalty value to a ranking value of the device rendered object; and
re-ranking, using the processor, the plurality of device rendered objects according to the penalty adjusted ranking value; and
transmitting, using the processor, the re-ranked plurality of device rendered objects to the client device for rendering via a display of the client device, wherein the plurality of device rendered objects is arranged for display according to the penalty adjusted ranking values of each device rendered object.
13. The method of claim 12 , wherein the penalty decision is generated using a logistic regression model.
14. The method of claim 12 , wherein the penalty value is generated using a machine learning model.
15. The method of claim 12 , wherein the ranking algorithm assigns a ranking value to each device rendered object of the plurality of device rendered objects based on one or more of users' preferences, the user's age group, the user's gender, the user's past purchases and clicks, popularity of objects, object attributes, object text title and descriptions, preference of the user for similar objects, distance between the object and the user.
16. The method of claim 15 , wherein the ranking algorithm comprises a plurality of ranking stages.
17. The method of claim 15 , wherein object attributes comprise price and inception date.