Training, inference and user interface for guiding the caching of media content on local stores

摘要

A system and method of caching multimedia data employing probabilistic predictive techniques is provided for facilitating local storage of a subset of available viewing selections by assigning a value to a selection and retaining selections in the cache depending on the value and size of the selection. The value assigned to an item can represent the time-dependent likelihood that a user (or group of users) will review an item at some time in the future, current viewing habits as well as habits over a particular time segment. A value assigned to a selection dynamically changes according to a set of cache retention policies, where the value can be time-dependent functions that decay based on the class of the item, for example. A selections value may also change based on changes in viewing habits, time segments or a user's modification of the cache retention policies.