Forecasting potential audience size and unduplicated audience size

A system for forecasting a potential audience size and an unduplicated audience size for a digital campaign, the system comprising:

• at least one memory including instructions;
• and
• at least one processor that is operably coupled to the at least one memory and that is arranged and configured to execute the instructions that, when executed, cause the at least one processor to implement a forecast module that is configured to:
• receive historic bid request data from a plurality of ad exchanges at a demand side platform (DSP);
• generate a plurality of time series models and a frequent itemset model based on the historic bid request data, wherein the frequent item set model is generated by iteratively applying a minimum threshold to attributes in a dataset and eliminating item sets that do not meet the minimum threshold;
• receive an audience segment input and a time period input, wherein the audience segment input comprises an OR clause;
• convert the audience segment input into a plurality of atomic target specifications by splitting the audience segment input based on the OR clause;
• for each of the plurality of atomic target specifications, determine a potential audience size during the time period input by:
• selecting a time series model from the plurality of time series models based on a frequency of attribute values from the corresponding atomic target specification,
• computing a conditional probability of the atomic target specification relative to a frequency of a segment corresponding to the time series model, and
• combining the selected time series model with the frequent item set model based on the conditional probability;
• aggregate the potential audience size for each of the plurality of atomic target specifications over the time period input into a total potential audience size;
• translate the total potential audience size to an unduplicated audience size by estimating a distribution of bid requests from unique cookies over a fixed time period in relation to the total potential audience size;
• and
• automatically respond to bid requests from the ad exchanges in real time using the DSP based at least in part on the total potential audience size and the unduplicated audience size.

A computer program product for forecasting a potential audience size and an unduplicated audience size for a digital campaign, the computer program product being tangibly embodied on a non-transitory computer-readable storage medium and comprising instructions that, when executed by at least one computing device, are configured to cause the at least one computing device to:

• receive a minimum threshold, a number of attributes in a transaction and a dataset including historic bid request data from a plurality of ad exchanges at a demand side platform (DSP);
• create a frequent item set model by iteratively applying the minimum threshold to the number of attributes in the dataset and eliminating item sets that do not meet the minimum threshold;
• receiving an audience segment input and a time period input, wherein the audience segment input comprises an OR clause;
• converting the audience segment input into a plurality of atomic target specifications by splitting the audience segment input based on the OR clause;
• for each of the plurality of atomic target specifications, determine a potential audience size during the time period input by:
• selecting a time series model based on a frequency of attribute values from the corresponding atomic target specification,
• computing a conditional probability of the atomic target specification relative to a frequency of a segment corresponding to the time series model, and
• combining the selected time series model with a frequent item set model based on the conditional probability;
• aggregate the potential audience size for each of the plurality of atomic target specifications over the time period input into a total potential audience size;
• translate the total potential audience size to an unduplicated audience size by estimating a distribution of bid requests from unique cookies over a fixed time period in relation to the total potential audience size;
• and
• automatically respond to bid requests from the ad exchanges in real time using the DSP based at least in part on the total potential audience size and the unduplicated audience size.