One or more computing devices, systems, and/or methods are provided. A machine learning model may be trained using a plurality of sets of information. One or more pruning operations may be performed in association with the training to generate a machine learning model with sparse vector representations associated with features of the plurality of sets of information. A request for content associated with a client device may be received. A set of features associated with the request for content may be determined. A plurality of positive signal probabilities associated with a plurality of content items may be determined using the machine learning model based upon one or more sparse vector representations, of the machine learning model, associated with the set of features. A content item may be selected from the plurality of content items for presentation via the client device based upon the plurality of positive signal probabilities.

A control system determines whether a browser is retrieving supplemental content over a slow or faulty network using a tracking indicator. Based on determining that browser is retrieving supplemental content over a slow or faulty network, the control system modifies responses to opportunities to provide supplemental content to the browser. The control system may modify responses by declining to bid, responding with a lower bid amount, or responding with alternate supplemental content. Based on determining that browser is no longer retrieving supplemental content over a slow or faulty network, the control system modifies responses to opportunities to provide supplemental content to the browser by responding with a higher bid amount or responding with the original supplemental content. Thus, the control system maximizes the use of control system and campaign manager resources by responding to opportunities according to the likelihood that they will result in an impression or a conversion.

Disclosed is delivering and closing out viewership priced spots in zoned cable delivery of linear TV, assembling past measured viewership data within a zone for a TV spot in a program, by aggregating records and augmenting aggregated logs with viewers' demographic data. This includes running an automated auction for advertising, at five minutes or less before airtime, announcing the spot, providing the past viewership data augmented with demographic data, with bids accepted for a rate, accepting and selecting a winning bid, provisioning an ad corresponding to the winning bid rate for an operator to deliver to the zone, after selecting the bid and before the airtime. After airing the ad, included is processing collected delivery logs within ten minutes of receipt, calculating current measured viewership during the spot, calculating an extended price, and reporting to a winning bidder the extended price, within six hours of airing the ad.

An improved architecture including system and methods for online advertising placement that provide possibly defaulting advertisement tags the opportunity to serve an advertisement ahead of a lower value tag that is guaranteed to fill, resulting in higher CPMs (i.e., Cost Per Mille) for web publishers. The system and methods are configured to deterministically render an advertisement impression from a list of possibly defaulting advertisements in a JavaScript-enabled web browser. The knowledge of the complete outcome of such an “ad chain” at render-time significantly reduces complexity and latency in the supporting ad server. The system and method centers around a novel JavaScript approach to detect when an advertisement has been loaded but not defaulted. Additionally, the system and methods integrate the network and RTB demand channels by looking at all demand sources simultaneously and selecting the buyer from within the user's browser, and address predictive pricing to further enhance the online advertising placement process.

A system and method receive a first signal at one or more processors of a promotional system that indicates automatic detection of presentation of a commercial, determine one or more online advertising settings associated with at least one of a product, service, or entity of interest in the commercial that was detected, and determine a current setting for the one or more online advertising settings from a search engine system. The online advertising settings are associated with a website of interest of the at least one of the product, service, or entity of interest . The system and method also send a second signal to the search engine system that temporarily changes the one or more online advertising settings during a time period following detection of the presentation of the commercial.

A content stake offering system is disclosed. The content stake offering system includes a content stake offering module, comprising computer-executable code stored in non volatile memory, a processor, and a plurality of computing devices. The content stake offering module, the processor, and the plurality of computing devices are configured to receive a request to sell a stake of content, determine a value of the content, generate a stake offering based on the value of the content, and update the value of the content. Determining the value of the content includes transferring data of a piece of content between the plurality of computing devices, recording a content data, which corresponds to the transferred data of the piece of content, in a database chunk, hashing the database chunk into a hashed database chunk, and appending the hashed database chunk to a block on a blockchain.

The subject matter of this specification can be embodied in, among other things, a method that receives at a server a first request to rebroadcast a first sponsored content item that was displayed on the social network to the first user to a second user. A second request to display at least one sponsored content item to the second user when the second user accesses an electronic document associated with the social network is received at the server. The first sponsored content item is retrieved for rebroadcast from one or more sponsored content items indicated for rebroadcast to the second user. The first sponsored content item is output to the second user of the social network in response to the second request to display and without requiring the second user to select the first sponsored content item for display.

An online system receives offline conversion data in real-time from a third party system regarding an offline user, the offline conversion data indicating an action performed by the offline user and identifying information for the offline user. The online system identifies a local user matching the offline user based on the identifying information for the offline user. The online system stores the offline conversion for the identified local user. The online system determines an attribution to a sponsored content item for the offline conversion. The online system presents updated information regarding the offline conversion to the third party system in real-time.

A media presentation and distribution system communicatively coupled to a client device is provided. The client device receives a media stream from the media presentation and distribution system via a network. The media stream includes media content and one or more tags. The client device further identifies in the received media content, the one or more tags, which corresponds to the overlay-graphic items. The client device further identifies candidate time-periods in the media content based on the one or more tags in the media content. At least one presentation attribute for the overlay-graphic items is identified based on the one or more tags in the media content. The client device further presents the overlay-graphic items at the candidate time-periods on the media content based on a degree of correlation between the identified tag in the media content and a corresponding context for the media content at the candidate time-period.

A method and system for enriching bid requests for real-time bidding on a digital advertisement placement are provided. The method comprises processing a received bid request to extract at least one data point, wherein the bid request is received from a website requesting placement of a digital advertisement; causing generation of at least one enriched data point based on the at least one extracted data point; and associating the at least one enriched data point with the received bid request to allow a refined real-time bidding on a placement of a digital advertisement in a webpage of the website in response to the bid on the received bid request.