Abstract: A creative is uniquely and optimally customized for every user impression, using the materials and tools available to those wishing to send image- and text-based messages in the market dominated by walled garden platforms, together with a creative e engine process that combines supervised machine learning, relational databases, and generative adversarial networks in a particular configuration that generates the creative. In a first phase, the engine uses machine learning to identify Creative visual features that are associated with high (or low) levels of Performance Metrics when included in Creatives served to Users with a given high-dimensional set of User Attributes. In the second phase, the engine automatically composes Creatives that are composed of visual features that are optimized to create high Performance Metrics when served to Users with a set of attributes that are determined in real time.

Abstract: A system and method for accurately matching corresponding DSP event data and Ad-Server event data with associated with a single real-world ad serve event by (a) pairing DSP event data and Ad-Server event data into data pairs, (b) comparing various field data in associated source fields from each of the DSP event data and Ad-Server event data to determine if the field data is a match or unmatch, and (c) based on the likelihood that a match of field data in a particular source field indicates an overall event match, which is determined using a Bayesian analysis, determining the probability that the DSP event data and Ad-Server event data in the data pair truly corresponding to the same single real-world ad serve event.

Abstract: A system and method for frequency capping across multiple DSPs based on a global frequency cap for all DSPs, wherein a particular DSP determines whether a target user has reached a global frequency cap across all DSPs based on (a) the actual number of ad displays the particular DSP has delivered to the target user and (b) the estimated number of ad displays that have been delivered to the user by the other DSPs, which is estimated using a Markov chain and Poisson probability analysis that determines the probability the target user is in each of a plurality of DSP universes and the likely rate of ads delivered in each universe to find the amount of time spent in each universe and the ads seen in each universe.

Abstract: A system and method for accurately matching corresponding DSP event data and Ad-Server event data with associated with a single real-world ad serve event by (a) pairing DSP event data and Ad-Server event data into data pairs, (b) comparing various field data in associated source fields from each of the DSP event data and Ad-Server event data to determine if the field data is a match or unmatch, and (c) based on the likelihood that a match of field data in a particular source field indicates an overall event match, which is determined using a Bayesian analysis, determining the probability that the DSP event data and Ad-Server event data in the data pair truly corresponding to the same single real-world ad serve event.