LOCALIZATION OF QOE MODELS TO ACCOUNT FOR USER PREFERENCES AND BIASES
Abstract:
In one embodiment, a device obtains quality of experience metrics specified by a plurality of users of an online application and user information regarding those users. The device identifies, based on the user information, one or more user characteristics of the plurality of users associated with variations in the quality of experience metrics. The device trains a prediction model to predict quality of experience metrics for a subset of the plurality of users based on the one or more user characteristics associated with the subset. The device makes, based on a prediction by the prediction model, an adjustment to a configuration of a network used by the subset to access the online application.
Information query
Patent Agency Ranking
0/0