Abstract: In some embodiments, a method receives a network trace from a playback session of a delivery of content being managed by a content management system. The network trace is decomposed into a first sub-trace and a second sub-trace based on a threshold. The threshold is used to divide the network trace into the first sub-trace and the second sub-trace. A value is extracted for a feature for the network trace based on the first sub-trace. The method classifies the network trace into a category based on the value and performs an action based on the category for the content management system.

Abstract: In some embodiments, a method determines whether to use first logic or second logic to select a profile from a list of profiles to download a segment of content. When using the second logic: selecting historical information from one or more previous playback sessions; determining a bandwidth using the historical information; and selecting the profile using a bitrate that is determined based on the bandwidth. When using the first logic, selecting the profile without using the historical information.

Abstract: In some embodiments, a method receives quality metric values for profile levels in a plurality of profile levels for a segment of content. An available bandwidth and associated bitrates of profile levels in the plurality of profile levels are evaluated to select a first profile level. The quality metric values for profile levels in the plurality of profile levels are evaluated to select a second profile level. The method selects a profile level in the plurality of profile levels based on the first profile level and the second profile level. The profile level that is selected is requested for the segment.

Abstract: In some embodiments, a method receives a data structure for an auto decision process for predicting a value for a decision feature of a plurality of features. The data structure is generated using a decision aware splitting process that splits a node in the data structure using the decision feature and a metric-based process that splits other nodes of the data structure by evaluating features in the plurality of features using a metric. Feature values are determined for a combination of features for a playback session. The method evaluates the data structure using the combination of features to determine a plurality of prediction values for a plurality of feature values of the decision feature. The prediction values are used to generate a decision for the playback session by selecting a feature value in the plurality of feature values based on the plurality of prediction values.

Abstract: In some embodiments, a method receives a request for an instance of media content in a current session. A state is retrieved for content delivery networks that is determined for a session in a set of sessions. The state includes a first set of values for a performance metric, and the first set of values is associated with the content delivery networks. The method determines a second set of values for the performance metric for the current session. The second set of values is associated with the content delivery networks. A content delivery network is selected for the request in the current session based on treating the current session as a session in the set of sessions. The session for the current session is after the session for the state in the set of sessions.

Abstract: In some embodiments, a method sends information for a sample of content, a first question, and a second question for output on an interface. The first question receives, from a subject, a first response for a sample level rating for an artifact that is perceived to be visible in the sample and the second question receives, from the subject, a second response for regions in the sample that are perceived to contain the artifact. The method receives the first response for the sample level rating and the second response for regions that are perceived to contain the artifact. First responses are combined from multiple subjects to generate an opinion score for the sample and second responses are combined to generate region scores for regions. The method generates training data from the opinion score and the region scores to train a process to perform an action based on the artifacts.

Abstract: In some embodiments, a method determines an input value for a request for content and retrieves a plurality of states for a plurality of content delivery networks. The state includes a component that is based on variance. A plurality of scores for a plurality of content delivery networks is generated based on the input value and the plurality of states. A score for a content delivery network is generated using a parameter value for the component and the input value. The method selects a content delivery network from the plurality of content delivery networks based on the score for the content delivery network and outputs information for the selected content delivery network to cause the selected content delivery network to service the request to deliver the content.

Abstract: In some embodiments, a method determines an input value for a request for content and retrieves a plurality of states for a plurality of content delivery networks. The state includes a component that is based on variance. A plurality of scores for a plurality of content delivery networks is generated based on the input value and the plurality of states. A score for a content delivery network is generated using a parameter value for the component and the input value. The method selects a content delivery network from the plurality of content delivery networks based on the score for the content delivery network and outputs information for the selected content delivery network to cause the selected content delivery network to service the request to deliver the content.

Abstract: In some embodiments, a method receives a request for an instance of media content in a current session. A state is retrieved for content delivery networks that is determined for a session in a set of sessions. The state includes a first set of values for a performance metric, and the first set of values is associated with the content delivery networks. The method determines a second set of values for the performance metric for the current session. The second set of values is associated with the content delivery networks. A content delivery network is selected for the request in the current session based on treating the current session as a session in the set of sessions. The session for the current session is after the session for the state in the set of sessions.

Abstract: In some embodiments, during playback of a video, using a parameter value of an adaptive bitrate algorithm to analyze playback of the video. The adaptive bitrate algorithm uses the parameter value to select a profile from a plurality of profiles to use to request segments of the video. A method selects a scenario in a plurality of scenarios when a playback condition meets a characteristic value of the scenario. The plurality of scenarios are associated with a different characteristic value. The method adjusts the parameter value of the adaptive bitrate algorithm based on the scenario to generate an adjusted parameter value. The plurality of scenarios have at least one different parameter value. The adjusted parameter value of the adaptive bitrate algorithm is used to analyze the playback of the video.