| CPC G06Q 30/0277 (2013.01) [G06F 16/2237 (2019.01); G06F 16/248 (2019.01); G06F 16/2455 (2019.01)] | 20 Claims |
|
1. A computer-implemented method, comprising:
assigning, by one or more processors, without user intervention and to each of a plurality of digital content objects, a corresponding set of one or more pseudo-randomly generated parameter values, each parameter value of the corresponding set of one or more pseudo-randomly generated parameter values corresponding to a parameter representing a feature of the plurality of digital content objects;
assigning, by the one or more processors, without user intervention and to each digital content object of the plurality of digital content objects, a weight parameter value indicating a likelihood of selecting the corresponding digital content object to provide to at least one client device for display as part of an exploration operation;
performing, by the one or more processors and without user intervention, the exploration operation to determine how to better at least one of tailor or rank online digital content, wherein the exploration operation comprises displaying one or more digital content objects to users via one or more client computing devices, tracking user interactions with respect to the one or more digital content objects and assigning score parameter values associated with the user interactions to at least some online digital content such that a plurality of impressions are substantially uniformly distributed among a set of digital content objects during the exploration operation, wherein the assigning score parameter values comprises calculating a score parameter value for each digital content object of the plurality of digital content objects based, at least in part, on:
the corresponding set of one or more pseudo-randomly generated parameter values;
additional one or more pseudo-randomly generated parameter values assigned to a query generated in response to online activity; and
the weight parameter value for the corresponding digital content object;
selecting, by the one or more processors, without user intervention and for display via a client device, at least one digital content object of the plurality of digital content objects based, at least in part, on one or more score parameter values determined based, at least in part, on the exploration operation performed to determine how to better at least one of tailor or rank online digital content; and
providing, by the one or more processors and without user intervention, the at least one digital content object for display by the client device, wherein at least one of the plurality of digital content objects or the set of digital content objects is stored in a distributed computing system employing a distributed process performed over a network of computing devices, wherein computational processing of at least one of the plurality of digital content objects or the set of digital content objects utilizing the distributed computing system is associated with efficiency in computational processing associated with at least one of selecting the at least one digital content object or providing the at least one digital content object for display.
|