A computer-implemented method may include predicting a first marketing indicator using a first module of an aggregate model; comparing the predicted first marketing indicator with a measured first marketing indicator; and based on the comparison of the predicted first marketing indicator with the measured first marketing indicator, adjusting the first module of the aggregate model. Additionally, the method may include predicting a second marketing indicator using a second module of the aggregate model; comparing the predicted second marketing indicator with a measured second marketing indicator; and based on the comparison of the predicted second marketing indicator with the measured second marketing indicator, adjusting the second module of the aggregate model. Further, the method may include determining a bid value based on the aggregate model.

Systems and methods for indirect advertising are disclosed. A real target is identified. Locations visited by the real target are identified. Secondary targets known to visit at least one of the identified locations are identified. A message designed to influence the real target is pushed to the secondary targets such that the real target may be influenced through word-of-mouth interaction with the secondary targets and led to believe that a larger phenomenon is occurring.

Disclosed are methods, systems, and non-transitory computer-readable medium for targeting bid and position for a keyword. For instance, the method may include obtaining information about the keyword, the information about the keyword including observations of value with respect to position for the keyword. The method may further include applying a Gaussian Process Model on the observations to obtain a prediction function and associated uncertainties, the prediction function and the associated uncertainties relating positions to expected values; applying a Thompson sampling reinforcement learning model on the expected values and the positions to obtain a target position; and applying a bid model to the target position to obtain bid information for the keyword. The method may also include transmitting a bid message to a search engine, the bid message including the bid information.

Aspects of the subject disclosure may include, for example, determining an ad play list for an Over-The-Top video stream requested by an end user device based on a price and/or a category of a creative derived from auctions with multiple Supply-Side Platform servers. The ad play list can be determined for the ad pod according to business rules, which can include a yield policy based on increasing revenue for a publisher of the video stream, and/or which can include a competitive separation policy enforced based on the category for bid responses. Other embodiments are disclosed.

Embodiments of the present invention provide for machine learning-based systems and methods for preventing fraud in programmatic advertising. The systems and methods provide for applying a plurality of machine learning models to data associated with a bid request, determining if the bid request is associated with fraudulent activity as a result of the machine learning models, and selectively preventing the bid request from being provided to potential buyers based on the determination.

A method for operating a server system. The method includes: receiving, by the server system, a request for a plurality of assets in a first format to fill an ad unit, in an application executing on a mobile device, with a native advertisement; receiving, by the server system and from an ad source over a network, ad content in a second format; converting, by a converter of the server system, the ad content in the second format to the plurality of assets in the first format; and sending, by the server system, the plurality of assets to the application, where the application fills the ad unit with the native advertisement including at least one of the plurality of assets.

Disclosed herein are systems and techniques for using a content delivery network to perform various functions within a digital advertising ecosystem, in ways that yield technological benefits such as improved security, efficiency, and speed (for example, reduction in publisher load times). As one specific example, a content delivery network can be used for the creation of electronic tokens for user identity protection between demand side platforms, supply side platforms, content creators (for example, advertisers), and publishers.

A computer-implemented method may include predicting a first marketing indicator using a first module of an aggregate model; comparing the predicted first marketing indicator with a measured first marketing indicator; and based on the comparison of the predicted first marketing indicator with the measured first marketing indicator, adjusting the first module of the aggregate model. Additionally, the method may include predicting a second marketing indicator using a second module of the aggregate model; comparing the predicted second marketing indicator with a measured second marketing indicator; and based on the comparison of the predicted second marketing indicator with the measured second marketing indicator, adjusting the second module of the aggregate model. Further, the method may include determining a bid value based on the aggregate model.

Embodiments of the present invention provide for real-time bidding of DOOH advertising units. The systems and methods provide for an API associated with at least one DOOH display, wherein the API is configured to receive a bid request indicating an available advertising unit from the at least one DOOH display, provide the bid request to at least one buyer; and receive a bid response from the at least one buyer, wherein the bid response includes content to be played at the at least one DOOH display.

A system and method capable of learning dynamic user and advertisement behavior for more effective click-through rate prediction. The system and method include at least one processor configured to obtain at least one item data, wherein the at least one item data comprises at least one explicit feedback to a user interaction event and other data associated with an item. The at least one processor also uses an interaction model that incorporates the obtained at least one item data to generate a user response prediction for a user and another interaction event