Mobile Content Delivery System with Recommendation-Based Pre-Fetching

• US 20140282636A1
• Filed: 10/24/2012
• Published: 09/18/2014
• Est. Priority Date: 10/24/2011
• Status: Abandoned Application

• Current Assignee: Nicta Ipr Pty. Ltd.
• Original Assignee: National ICT Australia Limited
• Inventors: Petander, Lars Henrik, Ott, Maximilian, Hitching, Robert, Maziere, Yorick Georges, Petterson, James, Webers, Christfried, Caetano, Tiberio Silva, Van de Van, Joop Johannes Wilhelmus, Noel, Joseph Christian Gruba
• Application Number:
  US14/350,643
• Publication Number:
  US 20140282636A1
• Time in Patent Office: Days
• US Class Current: 725/1
• CPC Class Codes: H04N 21/41407 embedded in a portable device, e.g. video client on a mobile phone, PDA, laptop constructional details of equipment or arrangements specially adapted for portable computer application G06F1/1626; arrangements specially adapted for mobile receivers in broadcast systems H04H20/57
• CPC Class Codes: H04N 21/422 Input-only peripherals , i.e. input devices connected to specially adapted client devices, e.g. global positioning system [GPS] input devices also receiving signals from specially adapted client devices H04N21/4104
• CPC Class Codes: H04N 21/44226 on social networks
• CPC Class Codes: H04N 21/458 Scheduling content for creating a personalised stream, e.g. by combining a locally stored advertisement with an incoming stream; Updating operations, e.g. for OS modules ; time-related management operations arrangements for replacing or switching information during the broadcast or during the distribution H04H20/10
• CPC Class Codes: H04N 21/4663 involving probabilistic networks, e.g. Bayesian networks

Abstract

This invention relates to the prefetching of content items, for example, prefetching of videos, to make the content items accessible from a local data store (118) of the mobile device (112). A processor (114) determines for each of the multiple content items a benefit value associated with an access time. The processor (114) also determines for each of the multiple content items an estimated prefetching cost associated with a prefetching time. Then, the processor (114) selects one of the multiple content items for downloading at the associated prefetching time based on the benefit values and the prefetching cost at the prefetching time. This way, the prefetching can be managed such that the most useful content item which the user will likely access soon are downloaded while at the same time minimising the cost.

Claims

1. A computer-implemented method for prefetching one of multiple content items onto a mobile device to make the content item accessible from a local data store of the mobile device, the method comprising:

(a) determining for each of the multiple content items a benefit value associated with an access time, the benefit value being based on an estimate of the likelihood of that content item to be accessed by a user at the access time;

(b) determining for each of the multiple content items an estimated prefetching cost associated with a prefetching time, the prefetching cost indicating the cost to download that content item at the prefetching time; and

(c) selecting one of the multiple content items for downloading at the associated prefetching time based on the benefit values for each of the multiple content items and the prefetching cost at the prefetching time for each of the multiple content items.

2. The computer-implemented method of claim 1, wherein step (c) comprises selecting the content item such that a combined measure of cost and benefit is optimised.

3. The computer-implemented method of claim 1, wherein step (a) comprises determining multiple benefit values for each of the multiple content items, each of the multiple benefit values being associated with one of multiple access times.

4. The computer-implemented method of claim 1, further comprising:

receiving historic network status data including time-based download costs; and

determining multiple prefetching times based on the historic network data, each prefetching time indicating a prefetching opportunity.

5. The computer-implemented method of claim 1, wherein the prefetching cost is based on a network policy.

6. The computer-implemented method of claim 5, wherein step (b) comprises determining multiple prefetching costs for each of the multiple content items, each of the multiple prefetching costs being associated with one of the multiple prefetching times.

7. The computer-implemented method of claim 6 wherein

step (b) comprises determining an earlier prefetching cost and a later prefetching cost for each of the multiple content items, the earlier prefetching cost being associated with an earlier prefetching time and the later prefetching cost being associated with a later prefetching time;

and

step (c) comprises selecting one of the multiple content items such that a first content item is selected over a second context item where the first content item has a combined measure of benefit and cost above a predetermined threshold associated with an access time before the later prefetching time and the second content item has a benefit value above a predetermined threshold associated with an access time after the later prefetching time.

8. The computer-implemented method of claim 6 wherein the prefetching cost is based on a monetary cost.

9. The computer-implemented method of claim 6, wherein the prefetching cost is based on an energy cost.

10. The computer-implemented method of claim 1, wherein step (a) further comprises:

receiving feature data including a set of features for describing content items and including for each feature of the set of features a feature weight associated with an access time;

receiving content data, including for each of the multiple content items one or more features of the set of features; and

determining for each of the multiple content items the benefit value based on the feature weight for each feature of the set of features.

11. The computer-implemented method of claim 10, further comprising

receiving for each feature and each context a feature context weight associated with that context, the feature context weight indicating how likely a feature is accessed by the user in the associated context;

receiving for each context a context weight associated with an access time, the context weight indicating how likely the context occurs at the associated access time; and

determining for each feature and for each context based on the context weight and the feature context weight a feature weight associated with an access time, the feature weight indicating how likely the user will access content items with that feature in each of the contexts.

12. The computer-implemented method of claim 11, further comprising:

receiving historic user access data including sensor data associated with access times when content items were accessed; and

determining contexts and context weights based on the sensor data, each of the context weights being associated with an access time.

13. The computer-implemented method of claim 12, wherein the historic user access data includes features of accessed content items and the method further comprises determining based on the received features and the determined contexts for each feature and for each context a feature context weight.

14. The computer-implemented method of claim 10, wherein the method further comprises:

receiving features of content items stored on the local data store; and

reducing the benefit value based on the received features.

15. The computer-implemented method of claim 1, wherein the benefit value is based on a creation date of the content item.

16. The computer-implemented method of claim 1, wherein one or more of the multiple content items are advertisement.

17. The computer-implemented method of claim 1, further comprising downloading the selected content item.

18. A non-transitory computer readable medium with an executable program stored thereon that when executed causes a computer to perform the method of claim 1.

19. A mobile device for prefetching one of multiple content items onto the mobile device to make the content item accessible from a local data store of the mobile device, the mobile device comprising a processor

(a) to determine for each of the multiple content items a benefit value associated with an access time, the benefit value being based on an estimate of the likelihood of that content item to be accessed by a user at the access time;

(b) to determine for each of the multiple content items an estimated prefetching cost associated with a prefetching time, the prefetching cost indicating the cost to download that content item at the prefetching time; and

(c) to select one of the multiple content items for downloading at the associated prefetching time based on the benefit values for each of the multiple content items and the prefetching cost at the prefetching time for each of the multiple content items.