| CPC G06Q 30/0275 (2013.01) [G06Q 30/0254 (2013.01); G06Q 30/0201 (2013.01)] | 20 Claims |
|
1. A computer-implemented method comprising:
receiving, by a service provider device and from a first customer device associated with device data, at least one purchase request signal that includes a customer identifier associated with a customer;
maintaining, by the service provider device, a product network, comprising:
a first node including a first product identifier representing a first product that was purchased by the customer, based at least in part on at least one first product purchase selection of at least one first product purchase link, the at least one first product purchase selection being received from the customer via at least one first user input to a first user interface (UI) of the first customer device;
a second node including a second product identifier representing a second product, wherein a plurality of first customers purchased both the first product and the second product via the service provider device, based at least in part on at least one second product purchase selection of at least one second product purchase link, the at least one second product purchase selection being received from at least one of the plurality of first customers via at least one second user input to at least one second UI of at least one second customer device; and
a third node including a third product identifier representing a third product, wherein a plurality of second customers purchased both the first product and the third product via the service provider device, based at least in part on at least one third product purchase selection of at least one third product purchase link, the at least one third product purchase selection being received from at least one of the plurality of second customers via at least one third user input to at least one third UI of at least one third customer device;
inputting, into a machine learning (ML) model, the product network and the device data;
determining, by the ML model and based at least in part on the product network and the device data, a first connection between the first product and the second product and a second connection between the first product and the third product;
outputting, by the ML model and based at least in part on the first connection being greater than the second connection, an interest level of the customer with respect to the second product;
calculating a recommended bid amount associated with the second product based at least in part on the interest level; and
transmitting, to a third party entity device associated with the second product and based at least in part on the recommended bid amount being greater than or equal to a threshold bid amount, a bid recommendation response including the recommended bid amount, the recommended bid amount and the device data being utilized by the third party entity device to transmit a bid message for causing presentation of content by a display of the first customer device, the recommended bid amount being included in a plurality of recommended bid amounts being transmitted to a first group of third party entity devices that includes the third party entity device, a first number of bid related computational tasks performed by the first group of third party entity devices being less than a second number of bid related computation tasks performed by a second group of third party entity devices, the second group of third party entity devices not receiving any recommended bid amounts.
|