Back to results
US 11,620,294 B2Granted
Dynamic media data management
AssigneePanasonic Avionics Corporation
InventorFaizalSheriff Kalifullah Sheriff
Life of this patent
Estimating the term…
Abstract
A system for managing media content data sets loaded onto a content server has a media content repository storing media content data items. A usage data aggregator retrieves usage log data from each passenger terminal device of the vehicle entertainment system. A load set generator builds a ranked media content data set based upon prioritization scores assigned to each of the media content data items with adaptive inferences from the usage log data. The prioritization scores are correlated with trip factors established for a subsequent vehicle trip for which the ranked media content data set is being built. A content retriever transfers the one or more media content data items specified in the ranked media content data from the media content repository.
Claims20
A system for managing media content data sets loaded onto a content server of a vehicle entertainment system presenting media content data items to passenger terminal devices, the system comprising:
- a media content repository storing one or more media content data items;
- a usage data aggregator in communication with the vehicle entertainment system, the usage data aggregator configured to:
- retrieve a first set of usage log data from a plurality of passenger terminal devices connected to the vehicle entertainment system, the first set of usage log data including first viewing preference and habit data associated with a plurality of passengers of at least one prior trip, the first set of usage log data further including demographic information of at least one of the plurality of passengers;
- and
- retrieve a second set of usage log data including second viewing preference and habit data from a subsequent trip that occurs after the at least one prior trip;
- a load set generator executing a machine learning model, the machine learning model trained using the first viewing preference and habit data including a destination and an origin of a prior trip to apply adaptive inferences to generate prioritization scores, the load set generator configured to:
- prior to retrieving the second set of usage log data:
- apply the machine learning model to one or more trip properties of the subsequent trip to generate the prioritization scores corresponding to respective media content data items, wherein the one or more trip properties include an origin or destination of the subsequent trip;
- assign the prioritization scores to each media content data item based upon the adaptive inferences from the first viewing preference and habit data, the prioritization scores being correlated with the one or more trip properties of the subsequent trip;
- and
- build a ranked media content data set for the origin and the destination of the subsequent trip based upon the assigned prioritization scores of each of the media content data items, wherein the ranked media content data set is stored in a table including a hash value corresponding to each of the media content data items that links the corresponding media content data item to a database including the first set of usage log data;
- and
- subsequent to retrieving the second set of usage log data:
- adapt the machine learning model using an aggregate of usage log data that includes the second set of usage log data;
- and
- a content retriever in communication with the media content repository configured to transfer the one or more media content data items specified in the ranked media content data set.
13 more elements
A media content manager for selectively loading media content data sets to an in-flight entertainment system, the media content manager comprising:
- a ground server configured to:
- receive a first set of usage log data including first viewing preference and habit data from passenger terminal devices connected to the in-flight entertainment system during a prior trip, the first viewing preference and habit data associated with a plurality of passengers of the prior trip, the first set of usage log data further including demographic information of at least one of the plurality of passengers;
- build a ranked media content data set based upon prioritization scores assigned to individual media content data items in the ranked media content data set, wherein the ranked media content data set is stored in a table including a hash value corresponding to each of the media content data items that links the corresponding media content data item to a database including the first set of usage log data,
- wherein the prioritization scores are generated by applying a machine learning model to one or more trip properties of a subsequent trip, wherein the one or more trip properties include an origin or destination of the subsequent trip, and wherein the machine learning model is trained by the first viewing preference and habit data to apply adaptive inferences to generate the prioritization scores for the subsequent trip,
- receive a second set of usage log data including second viewing preference and habit data from the subsequent trip, and
- adapt the machine learning model using an aggregate of usage log data that includes the second set of usage log data;
- and
- a content loader interface connectible to the ground server and a portable content loader device, the media content data items in the ranked media content data set being loaded on to the portable content loader device though the content loader interface.
5 more elements
A method for selectively loading media content data sets onto a vehicle entertainment system content server that presents media content items to passenger terminal devices, the method comprising:
- receiving, on a ground server, a first set of usage log data including first viewing preference and habit data from the passenger terminal devices connected to the vehicle entertainment system, the viewing preference and habit data being recorded by the passenger terminal devices during at least one prior trip, the first viewing preference and habit data associated with a plurality of passengers of the at least one prior trip, the first set of usage log data further including demographic information of at least one of the plurality of passengers;
- training a machine learning model using the first viewing preference and habit data, the machine learning model trained to apply adaptive inferences to generate prioritization scores;
- receiving, on the ground server, one or more trip properties of a subsequent trip;
- applying the trained machine learning model to the one or more trip properties to generate the prioritization scores corresponding to respective media content items, wherein the one or more trip properties include an origin or destination of the subsequent trip;
- assigning the prioritization scores to each media content item based upon the adaptive inferences from the viewing preference and habit data, the prioritization scores being correlated with the one or more trip properties;
- building a ranked media content data set based upon the assigned prioritization scores of each of the media content items, wherein the ranked media content data set is stored in a table including a hash value corresponding to each of the media content items that links the corresponding media content item to a database including the first set of usage log data;
- transferring the media content items in the ranked media content data set to the vehicle entertainment system content server;
- receiving, on the ground server, a second set of usage log data including second viewing preference and habit data from the subsequent trip;
- and
- adapting the machine learning model using an aggregate of usage log data that includes the second set of usage log data.
7 more elements
Dependencies are inferred from the claim text.