Rovi Guides, Inc.

Adeia Guides Inc.

Sciome, LLC

Kyle Miller

James W Lent

Bryan Stephen Scappini

Albert D Bennah

Jonathan B Gilpin

Ruchir Shah

Brian Edward Howard

Bryan S Scappini

Deepak Mav

Systems and associated methods are described for providing content recommendations. The system accesses a plurality of recommendation algorithms and assigns a plurality of weight values to each prediction algorithm. Then, the system generates a set of candidate weight combinations, such that each candidate combination includes a weight value assigned to each prediction algorithm. Then requests for content items are received over a predetermined period of time. For each combination, the system generates a set of recommended content items and an evaluation metric that is based on matches with requests. Afterwards, the system replaces a candidate combination that resulted in a generation of a lowest evaluation metric. The aforementioned steps are repeated until the evaluation metrics stop improving. Then display identifiers are displayed for a set of recommended content items generated for a candidate combination with the highest evaluation metric.

Systems and methods for providing media content recommendations

Systems and methods are described that employ machine learning models to optimize database management. Machine learning models may be utilized to decide whether a new database record needs to be created (e.g., to avoid duplicates) and to decide what record to create. For example, candidate database records potentially matching a received database record may be identified in a local database, and a respective probability of each candidate database record matching the received record is output by a match machine learning model. A list of statistical scores is generated based on the respective probabilities and is input to an in-database machine learning model to calculate the probability that the received database record already exists in the local database.

Automated metadata asset creation using machine learning models

Systems and associated methods are described for providing content recommendations. The system selects a first plurality of subsets of content categories, each subset of content categories comprising a first number of content categories. The subsets are assigned reward scores based on content popularity and duplication. The subset are then iteratively modified to increase the rewards scores. If the reward scores are still low, the process is repeated by selecting a second plurality of subsets of content categories, each subset of content categories comprising a second number of content categories, different from first number.

Evolutionary parameter optimization for selecting optimal personalized screen carousels

Systems and methods are described for selecting content item identifiers for display. The system may identify a set of content items that are likely to be requested in the future based on a history of content item requests. The system then selects a first plurality of content categories using a category selection neural net and selects a first set of recommended content items for the first plurality of content categories. The system increases a reward score for the first plurality of content categories based on receiving a request for a content item that is included in the first set of recommended content items. The system also decreases the reward score for the first plurality of content categories based on determining that the requested content item is included in the set of content items that are likely to be requested in the future. The neural net is trained based on the reward score of the first plurality of content categories to reinforce reward score maximization. The trained neural net is the used to select content items for display.

Deep reinforcement learning for personalized screen content optimization

Methods and systems for presenting contextually relevant push content when a user is passively engaged with an application are described herein. The system detects that the user is engaged with an application on a device and monitors the user's level of engagement with the application. If the system determines that the user is passively engaged, the system prepares to insert push content into the current output. The system identifies a region on the current output into which to insert push content, for example a region unoccupied by content or a particular object. The system identifies a context of the current output and selects a push content item based on the context of the current output. Then, the system inserts the push content item into the empty region on the current output.

Systems and methods for pushing content

Systems and associated methods are described for providing content recommendations. The system receives a plurality of sets of range values, each of which corresponds to a respective one of a plurality of parameters for recommending the content. The system selects different values within each of the plurality of sets of range values over time and provides a plurality of content recommendations to users based on the selected different values. The system then analyze users' behavior in response to the provided plurality of content recommendations. The system further updates at least one set of range values based on the analyzed users' behavior.

Systems and methods for providing content recommendations

Systems and associated methods are described for providing content recommendations. The system selects, using a multi-armed bandit solution model, a first plurality of content categories based on a reward score of each content category. The categories are displayed. When a user selects an item from the displayed categories, the system finds all categories that include the selected item, but rewards only the category with the highest score. The system selects, using the multi-armed bandit solution model, the second plurality of content categories based on the updated reward score of each content category. The categories are then displayed. The system may also repeat the steps to refine the multi-armed bandit solution model.

Methods and systems for personalized screen content optimization

Systems and methods are described that employ machine learning models to optimize database management. Machine learning models may be utilized to decide whether a new database record needs to be created (e.g., to avoid duplicates) and to decide what record to create. For example, a match machine learning model may be used in combination with an in-database machine learning model and an out-of-policy machine learning model. The match machine learning model may calculate a probability that a newly received database record individually matches respective candidate database records identified in a local database, the in-database machine learning model may calculate a probability that the received database record is in the local database, and the out-of-policy machine learning model may be used to calculate a probability that the newly received database record complies with inclusion rules of the local database.

Growing community of inventors

Durham, NC, United States of America

Kyle Miller