Apparatuses, methods, and systems for determining winning arms of electronic testing. One method includes obtaining historical data values related to the A/B test of a user, storing the historical data values, determining a historical weight for the historical data values, receiving new data values from the plurality of computing devices collected based on recipient actions during execution of the A/B, constructing a Dirichlet distribution, inferring corresponding central tendencies of samplings of a metric distribution, wherein each central tendency of the corresponding central tendencies is determined by sampling the Dirichlet distribution, constructing an overall utility distribution for each arms of the A/B test by combining the central tendency of each sampling of the metric distribution with a corresponding sampling of a conversion probability distribution, determining a winning arm of the A/B testing by comparing the overall utility distribution of each arm with each other arm of the A/B test.

Some aspects relate to technologies providing a framework for integrating two machine learning models to determine an attribution of a content campaign to conversions. In accordance with some aspects, a first machine learning model (such as a media mix modeling model) generates a first attribution of a content campaign to intermediate events. A second machine learning model (such as a multi-touch attribution model) generates a second attribution of the intermediate events to conversions. An attribution of the content campaign to the conversions is determined as a function of the first attribution and the second attribution.

A system and method for delivering promotional rewards to a customer mobile device comprising a promotional reward wheel, a media server, and messaging application operating on the customer's mobile device. A customer interacts with the promotional reward wheel and the system captures that interaction and responsive to the promotional reward that was won by the interaction, generates a message on the customer's mobile device with a means to claim the reward by sending the generated message to the appropriate message endpoint.

A method of determining effectiveness of an advertising campaign comprising: bidding on available advertising inventory; obtaining online impression data from at least one advertisement for at least one successful bid; matching, via a data cloud, the online impression data to a plurality of unique consumers; matching, via the data cloud, a plurality of unique consumers to at least one unsuccessful bid; determining characteristics of consumers for impression data who performed a desired event; determining characteristics of consumers for impression data who did not perform the desired event; determining characteristics of consumers from the at least one unsuccessful bid who performed the desired event; determining characteristics of consumers from the at least one unsuccessful bid who did not perform the desired event; determining consumer characteristics likely to lead to the desired event; and measuring the impact of the at least one successful bid.

Dynamic campaign optimization systems and methods may be used to continuously test many alternative campaign configurations while allowing all configurations, including configurations formerly identified as successful and unsuccessful, to be re-tested in order to identify successful configurations that may previously have been identified as unsuccessful.

Systems and methods for optimizing selection of media content for long-term outcomes are provided. Observational data including intermediate outcomes from an observation period is combined with historical data to select media content based on estimated long-term outcomes at the end of an optimization period. As time passes, the observational data is updated with more intermediate outcomes, allowing more accurate estimates of long-term outcomes to be made. In an example, a predictive model trained using the historical data uses the observational data to estimate distributions of long-term outcomes. An action selector selects samples from the distributions and selects media content based on the samples.

In one embodiment, a method includes obtaining, from a demand-side platform (DSP), impression data specifying service providers and consumer tokens representing consumers who have received digital impressions of a set of advertising campaigns. A set of tokenized claims data records related to a prescription of a product is then received from a database server. A result set of integrated measurement records specifying measured campaigns linking the tokenized claims data records with impression data associated with consumer tokens and/or service provider identifiers is further received from the database server. Aggregated analytics reports based on the integrated measurement records are generated and presented. A machine learning model may be trained using a training dataset comprising features selected from the impression data and tokenized claims data records, to predict bid values or other parameters for use in updating, optimizing or modifying operation of the DSP for the original campaign or for other campaigns.

In accordance with some embodiments of the disclosed subject matter, methods, systems, and media for setting and using an advertisement frequency cap based on causal conversions or impact of advertisements are provided.

Systems, methods and media for adaptive real time modeling and scoring are provided. In one example, a system for automatically generating predictive scoring models comprises a trigger component to determine, based on a threshold or trigger, such as a detection of new significant relationships, whether a predictive scoring model is ready for a refresh or regeneration. An automated modeling sufficiency checker receives and transforms user-selectable system input data. The user-selectable system input data may comprise at least one of email, display or social media traffic. An adaptive modeling engine operably connected to the trigger component and modeling sufficiency checker is configured to monitor and identify a change in the input data and, based on an identified change in the input data, automatically refresh or regenerate the scoring model for calculating new lead scores. A refreshed or regenerated predictive scoring model is output.

Executing a machine learning model in an artificial intelligence infrastructure that includes one or more storage systems and one or more graphical processing unit (GPU) servers, including: receiving, by a graphical processing unit (GPU) server, a dataset transformed by a storage system that is external to the GPU server; and executing, by the GPU server, one or more machine learning algorithms using the transformed dataset as input.