Systems and methods are disclosed herein for a recommendations engine that generates content recommendations using a trained model that is personalized based on the information corresponding to content consumption. The disclosed techniques herein provide a trained model to provide content recommendations. The trained model may have been trained using a predefined set of training data agnostic of a particular user profile. A system receives information corresponding to content consumption. The system may associate the information corresponding to content consumption with a profile. The system generates a personalized model based on the information corresponding to content consumption and on the trained model. The personalized model may be associated with the user profile. The system generates the content recommendations using the personalized model. The system then causes to be provided the content recommendations.

Embodiments include systems and methods for analyzing content of a media presentation within a receiving computing device. In embodiments, a processor of a computing device may analyze media data within a media bitstream while receiving, such as prior to or in parallel with presentation of, the media data to determine whether the media data includes objectionable content. The processor may modify a presentation of the media data in response to determining that the analyzed media data includes objectionable content. The computing device may be a wireless communication device.

Data processing systems and methods are disclosed for combining video content with one or more augmentations to produce augmented video. Objects within video content may have associated bounding boxes that may each be associated with respective RGB values. Upon user selection of a pixel, the RGBA value of the pixel may be used to determine a bounding box associated with the RGBA value. The client may transmit an indicator of the determined bounding box to an augmentation system to request augmentation data for the object associated with the bounding box. The system then uses the indicator to determine the augmentation data and transmits the augmentation data to the client device.

Exemplary embodiments are directed to a device and method for controlling ambiance based on user-requested content. The device receives training data, user content requests, and user network device modification data. The device ascertains a content type for a user content request, performs analytics on the training data to create a model to predict ambiance settings, predicts ambiance settings for the content type using the model, evaluates the accuracy of the model and performs optimizations to the model to improve the predictions of the ambiance settings. The device controls the operating settings of one or more network devices based on the predicted ambiance settings. Moreover, exemplary embodiments are directed to controlling the enabling of display notifications, controlling reminder notifications, and controlling greeting notifications using a face identifier and notification settings.

Systems and methods for generating and presenting content recommendations to new users during or immediately after the onboarding process, before any history of the new user's viewed content is available. A machine learning or other model may be trained to determine clusters of content genre values corresponding to genres of content watched by viewers. Clusters are thus associated with popular groupings of content genres viewed by many users. Clusters representing popular groupings of content genres may be selected for new users, and content corresponding to the selected clusters may be recommended to the new users as part of their onboarding process. A sufficient amount of content may be selected to fully populate any content recommendation portion of a new user onboarding page.

Methods and systems for determining and using disliked content are described. The method includes obtaining, by a service provider system, channel viewing data from a user device of a user, pre-processing, by a content analysis unit, the channel viewing data to mitigate inaccuracies and inconsistencies in the channel viewing data, scaling, by the content analysis unit, the pre-processed channel viewing data to highlight temporal dislike aspects of the pre-processed channel viewing data, applying, by the content analysis unit, machine learning algorithms to the scaled channel viewing data to generate a disliked content ratings matrix, and outputting, by the content analysis unit to user interface systems in the service provider system, to provide enhanced user viewing. The user interface systems including a recommender system, an alternate content generation system, or both.

Various exemplary embodiments include a valuation system to measure sponsorship exposure in social media, including the ability to use multiple valuation methods, such as CPV, CPM, and CPE, integrated into the valuation system, in a real-time manner, while aggregating the most up-to-date data. It may categorize the social media by media type, image, video or text, and may run different valuation strategies based on the media type. It may adapt a valuation strategy based on the source platform of the publication, detect the sponsorship exposures and inputs into the valuation method, as one or more factors to the valuations system, in the real-time matter. Additionally, granular valuation may be supported, given the output from the AI-driven system, on brand, asset, scene, and media exposure types. It also supports valuation on a real-time ad rate, a device factor, customization based on user configurations, e.g., discounted factor, and/or supports live stream.

Prediction models for managing viewership data are disclosed. An amount of time users are displayed content is initially obtained. The obtained amounts may be for each content distributor that distributes channels, for each of the channels with respect to which sets of content are displayed, for each of the sets that comprises content displayed during past periods, and for each of the displayed content. A set of features associated with each of the displays is obtained and a target period is selected from among the past periods. A model is used to predict an amount of time users were displayed content during the target period, for each of the sets, channels, and distributors, based on the obtained sets of features associated with the displays during the target period and on the obtained amounts for the displays during the past periods that precede the target period. A comparison, respectively for the same displays during the target period, of each of the obtained amounts to each of the predicted amounts is performed, and an anomaly is detected based on the comparisons. Finally, the anomaly is alerted.

Systems, methods, and non-transitory computer-readable media can determine a broadcaster request to determine information for conducting a content broadcast through the computing system. One or more parameters for the broadcast can be determined using a machine learning model that has been trained to predict the one or more parameters based at least in part on data describing previously conducted broadcasts. Information that describes at least the one or more parameters is provided to the broadcaster.

Techniques are disclosed for characterizing audience engagement with one or more characters in a media content item. In some embodiments, an audience engagement characterization application processes sensor data, such as video data capturing the faces of one or more audience members consuming a media content item, to generate an audience emotion signal. The characterization application also processes the media content item to generate a character emotion signal associated with one or more characters in the media content item. Then, the characterization application determines an audience engagement score based on an amount of alignment and/or misalignment between the audience emotion signal and the character emotion signal.