Abstract: A method includes receiving pixel data from a universal pixel embedded in a website and exposure data from a publisher system, enriching the exposure data with program metadata to generate enriched exposure data, classifying identifier types present in the pixel data into different categories, matching engagement data against the exposure data based on the different categories of identifiers, and generating an attribution report identifying which targeted data delivery exposure led to a website conversion event.

Abstract: A method includes training multiple models in parallel based on model training requirements and training data received from one or more external data systems, evaluating candidate models from among the trained data models based on feedback data received from a bidder system to generate performance metrics for the candidate models, and selecting an optimal data model for each line item based on the performance metrics.