Systems and methods for constraint-based optimization, comprising: an AI demand forecasting engine, an optimization engine, a user-defined objective, and a user-defined set of constraints. Using historical sales data, the AI demand forecasting engine generates a plurality of entities, each entity defined by a placement of an item in a promotion platform; and forecasts the objective associated with each entity. The optimization engine generates a plurality of plans, each plan consisting of a unique subset of entities. Plans that violate at least one constraint are eliminated by the optimization engine, leaving a set of candidate solutions. An optimum plan is selected from the set of candidate solutions based on maximization of the objective.

In some examples, a computing device may receive data from a plurality of groups of data sources. The computing device may create a training data set from a first portion of the received data and may create a plurality of validation data sets from a second portion of the received data. For example, each validation data set may correspond to a respective one of the groups of data sources. The computing device may train, using the training data set, a plurality of machine learning models configured for synthesizing data. For instance, respective ones of the machine learning models may correspond to respective ones of the groups of data sources. Further, the computing device may validate the respective machine learning models using the respective validation data set corresponding to the respective group to which the respective machine learning model being validated corresponds.

A computer-implemented method for sending email messages with dynamically-determined content, the method comprising: transmitting email messages to a plurality of recipient groups each comprising at least one recipients, each message comprising one or more dynamic content network references associated with dynamically-selected content items; receiving, by a network-connected server, a dynamic content network request responsive to the rendering of one of said dynamic content network references within an email message received by a recipient device, the dynamic content network request comprising one or more content selection criteria, the content selection criteria comprising an embedded user identifier uniquely associated with a user to whom an email among the email messages was transmitted; dynamically selecting a content item from amongst a plurality of candidate content items based at least in part on the content selection criteria and the recipient's recipient group; implementing an exclusion service by cross-referencing a dynamic content item with static content associated with the embedded user identifier and, based on a determination that the dynamic content item is duplicative of an associated static content item within a same email communication, serving an alternative non-duplicative content item in response to the dynamic content request; returning the selected or the alternative content item to the email recipient device for rendering within the same email communication; and determining relative effectiveness of content by recipient group.

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.

In one embodiment, a method includes accessing a current-visitor context of a current visitor to a web page in a current web-browsing session. The current-visitor context includes one or more data associated with or concerning the current visitor. The method includes selecting based on the current-visitor context a particular one of multiple possible instances of the web page for presentation to the current visitor. The particular one of the multiple possible instances of the web page is substantially most likely to generate a highest expected outcome from interaction with the web page by the current visitor as indicated by the current-visitor context.

A digital brand asset system is provided enabling a brand owner to create, distribute, maintain, manage, merchandise and analyze smart brand assets. The system enables distribution and sharing of smart brand assets across the websites. The websites can host webpages containing codes representing the smart brand assets. When a user device retrieves a webpage from one of the websites and renders the webpage, it executes the codes and requests the content of the smart brand assets from a brand asset server. Through the brand asset server, a brand owner can control the content and the presentation of the smart brand asset hosted by the websites, based on various factors such as previous click through rates, aggregated shopper behaviors, geographical locations of the websites or website visitors, categorized types of websites, blacklist of websites.

A digital brand asset system is provided enabling a brand owner to create, distribute, maintain, manage, merchandise and analyze smart brand assets. Generally, the system enables distribution and sharing of smart brand assets across websites. The system performs the steps of presenting a web page containing code representing a smart brand asset that has a unique identifier, receiving a request for the smart brand asset from a search engine crawler which is indexing web pages of the web server, redirecting the request to a brand asset proxy server based on the unique identifier and satisfying the request by providing content of the smart brand asset. The unique identifier can include information of the location, user attributes, or the content of the smart brand asset. As a result, it is determined that the request is sufficiently satisfied to be indexed by the search engine.

Methods, apparatus, systems and articles of manufacture are disclosed to identify sponsored media in a Document Object Model. An example method includes in response to initiation of a search for sponsored media in a first branch of a document object model for a webpage, initiating a first timer having a first search time period and a second timer having a second search time period, the first search time period and the second search time period overlap, the search to be terminated upon completion of traversing the first branch when the search reaches a document node that is at a root of the first branch, the first branch including a candidate node. The example method also includes in response to identifying the candidate node as sponsored media, parsing the sponsored media to generate monitoring information, and modifying a user agent setting to include a unique identifier.

An on-line marketplace provides systems and methods for a mutually beneficial messaging campaign with one or more merchants, in order to promote the merchants' products. One aspect of the messaging campaign includes augmenting the merchants' budgets across multiple commercial channels. The channels may include merchant-operated channels, marketplace-operated channels, and/or third party channels. The marketplace selects when and how to augment the merchants' budgets. The augmented budget comprises an overall content budget for the messaging campaign, including a first budget portion provided by the merchant(s) and a second budget portion provided by the on-line marketplace. The augmented budget may be determined based on a weighted combination of goals of the marketplace and one or more merchants. The augmented budget may be segmented into static and dynamic portions, in which the dynamic portion is varied between channels to determine variables to optimize budget allocation and message performance.

In one embodiment, a method includes accessing a current-visitor context of a current visitor to a web page in a current web-browsing session. The current-visitor context includes one or more data associated with or concerning the current visitor. The method includes selecting based on the current-visitor context a particular one of multiple possible instances of the web page for presentation to the current visitor. The particular one of the multiple possible instances of the web page is substantially most likely to generate a highest expected outcome from interaction with the web page by the current visitor as indicated by the current-visitor context.