Abstract: One or more computing devices, systems, and/or methods for implementing a model for serving exploration traffic are provided. An amount of spend by a content provider to provide content items of the content provider through a content serving platform to client devices of users is determined. A number of exploration impressions of users viewing exploration content items of the content provider over a timespan is determined. A return on exploration impression metric is determined for the content provider based upon a ratio of the amount of spend to the number of exploration impressions. The return on exploration metric is used to rank available exploration content items of content providers for serving exploration traffic.

Abstract: A computer-implemented method for detecting and monitoring bias in an application includes index training data and obtaining a plurality of correlation values of one or more features in the indexed training data with a target variable. For each of the one or more features, a first value and a favorable result and a second value along with the unfavorable result is calculated. An absolute value of a difference between the calculated first value and the calculated second value is calculated. A total sum of the calculated absolute value of the plurality of correlation values of the one of the one or more features is calculated.

Abstract: Systems and methods for automatically aggregating, evaluating, and providing a contextually relevant offer or incentive is disclosed. One method obtains a plurality of offers for a customer. Each of the plurality of offers is then evaluated to determine a contextual relevance of each of the plurality of offers. Each of the plurality of offers is then sorted into a best contextually relevant offer to a least contextually relevant offer order. The obtaining, evaluating, and sorting is continuously repeated until a time for using an offer is identified. The system and method then present the best contextually relevant offer.

Abstract: A system and method for automatically predicting deviation on a metric of a use-case and deriving interconnections between metrics for generating action recommendations is provided. The system includes a deviation management system 104 which captures data from a plurality of external sources and internal sources and comprises of a deviation management platform 106 and a deviation management environment 108. The system includes various computation modules which work the deviation management platform 106 to provide a deviation management service to a set of clients that are associated with that service. The service and its users are specific to use-case, wherein the use-case is specified by a client device 116 inside the system. The system comprises of external data which is horizontal across a plurality of deviation management services and internal data which is specific to every deviation management service.

Abstract: Systems, methods, and apparatuses for storing verified identification information in a distributed database and for verifying entities to requestors are provided herein. Identity information relating to an entity is verified by a generally recognized identity provider (GRIP). Once verified, the GRIP populates a distributed database with an entry corresponding to the verified identity information. The distributed database is publically accessible such that identity requestors can see the entries in the distributed database. In some arrangements, the distributed database is write-protected such that only an authorized GRIP can modify the distributed database. An entity can provide information to a requestor to verify an identity attribute of the entity to the requestor. The requestor uses information provided by the entity to retrieve verified entity identification information from the distributed database and to confirm that the identity attribute was verified by an authorized GRIP.

Abstract: In an approach to hyperparameter optimization, one or more computer processors express a hyperparameter tuning process of a model based on a type of model, one or more dimensions of a training dataset, associated loss function of the model, and associated computational constraints of the model, comprising: identifying a set of optimal hyper-rectangles based a calculated local variability and a calculated best function value; calculating a point as a representative for each identified potentially optimal hyper-rectangle by locally searching over the identified set of potentially optimal hyper-rectangles; dividing one or more hyper-rectangles in the identified set of optimal hyper-rectangles into a plurality of smaller hyper-rectangles based on each calculated point; and calculating one or more optimal hyperparameters utilizing a globally converged hyper-rectangle from the plurality of smaller hyper-rectangles.

Abstract: Provided are a computer-implemented method, a system, and a computer program product. The method comprises extracting features from a plurality of base models in an ensemble model. The plurality of base models are configured to provide respective prediction results. The ensemble model is configured to provide an overall prediction result from the prediction results of the plurality of base models. The features are associated with time performance of the base models. The method further comprises clustering the plurality of base models into a plurality of clusters based on the extracted features. The method further comprises assigning the plurality of base models to a plurality of parallel computation units based on the plurality of clusters.

Abstract: Methods, apparatus, systems and articles of manufacture to estimate and deduplicate audiences are disclosed herein. An example apparatus includes a controller to determine a subunion of at least first and second marginal audiences of media based on of panel data and census data, the panel data including a panel impression count and a panel audience size, and the census data including a census impression count, an audience size calculator to determine a census audience size of the at least the first and second marginal audiences based on the panel impression count and the panel audience size and determine a subunion census audience size, the subunion census audience size corresponding to an overlap between the at least the first and second marginal audiences; and a report generator to generate a report including the census audience size and the subunion census audience size.

Abstract: A travel prediction method, apparatus, device, and storage medium.

Abstract: Systems, computer program products, and methods are described herein for generating an execution sequence using learning reinforcement. The present invention is configured to electronically receive one or more requirement modules from one or more computing devices associated with one or more resource development teams; store the one or more requirement modules in a resource development pipeline, wherein the one or more requirement modules are in a first order of execution; initiate a reinforcement learning algorithm on the one or more requirement modules; determine, using the reinforcement learning algorithm, a final order of execution for the one or more requirement modules; initiate a reordering of the one or more requirement modules from the first order of execution to a final order of execution; and initiate an execution of the one or more requirement modules in the final order of execution.

Abstract: Methods, apparatus, systems, and articles of manufacture for communication with an audience metering device are disclosed. An example apparatus includes one or more non-transitory computer readable media, instructions in the apparatus, and one or more processors to execute the instructions. The example one or more processors are to segment a message to be transmitted to a configuration device into a first message segment and a second message segment, store the first message segment in a characteristic memory, and transmit a first advertisement to the configuration device when the first message segment is stored in the characteristic memory. Additionally, the example one or more processors are to after the first message segment has been gathered by the configuration device, store the second message segment in the characteristic memory and transmit a second advertisement to the configuration device when the second message segment is stored in the characteristic memory.

Abstract: The present disclosure provides systems and methods configured to perform the stages of: receiving an initial device location; receiving a subsequent device location; determining a course of travel associated with the device; aggregating data associated with courses of travel for a plurality of devices; determining a plurality of commonly traveled paths based on the aggregated data; receiving an identification of a path of the plurality of commonly traveled paths; and defining content to be transmitted to at least one device traveling down the identified path.

Abstract: Systems and methods are for routing electronic payment transactions to PIN-less networks using payment pseudo-networks and electronic transaction simulation.

Abstract: Tiered advertisement bidding is disclosed. One or more quality metrics associated with a user profile are determined. An advertisement bid is selected from a plurality of tiered bids based at least in part on the determined quality metrics. Determining the quality metrics can include determining a conversion assessment. Determining the quality metric can also include determining whether a need associated with a user profile has been met for a category. In some cases, persona detection is performed with respect to the user profile.

Abstract: Systems, devices, media, instructions, and methods are provided for presentation of media collections with automated interactive advertising. In one embodiment, a client device receives content elements for display as part of a content collection. Advertising data is also received for display between selected content elements. Interaction elements are merged with the create an advertising element. During display of the advertising data, the interaction elements are presented on the client device output, and are controllable via user inputs. In various embodiments, interaction data recorded at the device is used to manage the presentation of future advertising data.

Abstract: This disclosure relates generally to categorical time-series clustering. In an embodiment, the method for categorical time-series clustering for categorical time-series associated with distinct subjects obtained from sensors. Based on the categorical time-series, the subjects are clustered into clusters by using a Markov chain model. Clustering the subjects include assigning each subject to a cluster. The subjects are assigned to the clusters by determining cluster-specific transition matrices based on a transitional probability of the subject's transitioning between states. A semi-distance function is constructed for each cluster-specific transitional matrix between the states at multiple time instances, which us indicative of a conditional probability of movement of the subject between the states at different time instance.

Abstract: Techniques and solutions are described for analyzing results of a machine learning model. A result is obtained for a data set that includes a first plurality of features. A plurality of feature groups are defined. At least one feature group contains a second plurality of features of the first plurality of features. The second plurality of features is less than all of the first plurality of features. Feature groups can be defined based on determining dependencies between features of the first plurality of features, including using contextual contribution values. Group contextual contribution values can be determined for feature groups by aggregating contextual contribution values of the constituent features of the feature groups.

Abstract: Methods, systems, and media for presenting an advertisement are provided. In some embodiments, the method comprises: causing an advertisement to be presented in connection with presentation of a video having content associated with an application, wherein the advertisement includes information indicating one or more virtual items associated with the application available for purchase; receiving a user input indicating a purchase of at least one of the one or more virtual items; causing payment information to be collected based on a user account associated with a request for presentation of the video to complete the purchase; and transmitting information indicating the completed purchase to an entity associated with the application.

Abstract: Methods, apparatus, systems, and articles of manufacture for communication with an audience metering device are disclosed. An example apparatus includes one or more non-transitory computer readable media, instructions in the apparatus, and one or more processors to execute the instructions. The example one or more processors are to segment a message to be transmitted to a configuration device into a first message segment and a second message segment, store the first message segment in a characteristic memory, and transmit a first advertisement to the configuration device when the first message segment is stored in the characteristic memory. Additionally, the example one or more processors are to after the first message segment has been gathered by the configuration device, store the second message segment in the characteristic memory and transmit a second advertisement to the configuration device when the second message segment is stored in the characteristic memory.

Abstract: Techniques disclosed herein relate generally to evaluating and selecting candidate datasets for use by software applications, such as selecting candidate datasets for training machine-learning models used in software applications. Various machine-learning and other data science techniques are used to identify unique entities in a candidate dataset that are likely to be part of target entities for a software application. A merit attribute is then determined for the candidate dataset based on the number of unique entities that are likely to be part of the target entities, and weights associated with these unique entities. The merit attribute is used to identify the most efficient or most cost-effective candidate dataset for the software application.