System and method for efficiently providing a recommendation

• US 8,548,987 B2
• Filed: 10/31/2008
• Issued: 10/01/2013
• Est. Priority Date: 09/24/1999
• Status: Expired due to Term

• Current Assignee: CDN Innovations, LLC (IP Investments Group LLC)
• Original Assignee: Thalveg Data Flow LLC (Intellectual Ventures LLC)
• Inventors: Rauser, John, Guralnik, Valerie
• Primary Examiner(s): TIMBLIN, ROBERT M
• Application Number:
  US12/263,048
• Publication Number:
  US 20090063439A1
• Time in Patent Office: 1,796 Days
• Field of Search: 707/719, 707/754
• US Class Current: 707/719
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

A system and method for applying a first filter and a second filter, such as a recommendation and a constraint filter, to a plurality of items, including determining a cost of applying the first filter and the second filter to the plurality of items, and determining an order of applying the first and second filters based on the cost of applying the first and second filters.

Claims

1. A method for applying filters to a plurality of items in a data processing system comprising:

receiving a recommendation request;

specifying a constraint filter that selects items satisfying a constraint;

determining an order of the filters based on a cost of applying the filters, including;

applying the constraint filter first when a cost of applying the filters when the constraint filter is applied first is lower than a cost of applying the filters when a recommendation filter is applied first, wherein the constraint filter determines if each item satisfies the constraint, and

applying the recommendation filter first when the cost of applying the filters when the recommendation filter is applied first is lower than the cost of applying the filters when the constraint filter is applied first, wherein the recommendation filter assigns a predicted enjoyment value to each item.

2. The method of claim 1, wherein the determining the order of the filters further comprises:

analyzing a cost for applying the filters to the items; and

determining a probability that the items will pass both filters.

3. The method of claim 1, wherein the cost of applying the filters is determined by the equation:

C=M/P(G+R),

wherein M is a number of desired items, P is a probability that the items will pass a second applied filter, G is a time to retrieve the items, and R is a time to decide if the items will pass the second applied filter.

4. The method of claim 1, further comprising:

calculating the cost of applying the filters based at least on a probability that a randomly selected item of the plurality of items will pass a second applied filter of the recommendation and constraint filters.

5. The method of claim 4, wherein the calculating is further based on a cost of applying a first applied filter of the recommendation and constraint filters to generate a single item.

6. The method of claim 4, wherein the calculating is further based on a cost of applying a second applied filter of the recommendation and constraint filters to a single item.

7. The method of claim 4, wherein the calculating is further based on a number of results required.

8. A method, comprising:

determining an order for applying a constraint filter and a recommendation filter to a plurality of items, comprising;

calculating a cost of applying the constraint filter and the recommendation filter in a first order based at least on

(i) a number of results required,

(ii) a probability that a randomly selected item of the plurality of items will pass a second applied filter of the constraint filter and the recommendation filter,

(iii) a cost of applying a first applied filter of the constraint filter and the recommendation filter to generate a single item, and

(iv) a cost of applying the second applied filter of the constraint filter and the recommendation filter to the single item; and

generating a recommendation list, comprising applying the constraint filter and the recommendation filter to the plurality of items according to the determined order, wherein the constraint filter determines if each item satisfies a constraint and the recommendation filter assigns a predicted enjoyment value to each item.

9. The method of claim 8, further comprising:

receiving a recommendation request.

10. The method of claim 8, further comprising:

displaying the recommendation list.

11. The method of claim 8, wherein the determining an order comprises calculating the cost using the equation:

cost=(number of results required/probability that a randomly selected item will pass the second applied filter)*(cost of applying the first applied filter to generate a single item+cost of applying the second applied filter to the single item).

12. A system, comprising:

a recommendation server comprising a processing device configured to;

receive a recommendation request,

determine an order for applying a constraint filter and a recommendation filter to a plurality of items based on a cost of applying the constraint filter and the recommendation filter, wherein the constraint filter determines if each item satisfies a constraint and the recommendation filter assigns a predicted enjoyment value to each item, and

generate a recommendation list, the generating comprising applying the constraint filter and the recommendation filter in the determined order.

13. The system of claim 12, wherein the cost is based at least on a probability that a randomly selected item of the plurality of items will pass a second applied filter of the constraint filter and the recommendation filter.

14. The system of claim 12, wherein the cost is based at least on a cost of applying a first applied filter of the constraint filter and the recommendation filter to generate a single item.

15. The system of claim 12, wherein the cost is based at least on a cost of applying a second applied filter of the constraint filter and the recommendation filter to a single item.

16. The system of claim 12, wherein the cost is based at least on a number of results required.

17. A computer program product comprising a non-transitory computer readable storage medium having control logic stored therein, the control logic, when executed, causing a processor to perform operations comprising:

receiving a recommendation request;

determining an order for applying a constraint filter and a recommendation filter to a plurality of items based on a cost of applying the constraint filter and the recommendation filter, wherein the constraint filter determines if each item satisfies a constraint and the recommendation filter assigns a predicted enjoyment value to each item; and

generating a recommendation list, the generating comprising applying the constraint filter and the recommendation filter in the determined order.

18. The computer program product of claim 17, wherein the cost is based at least on (i) a number of results required, (ii) a probability that a randomly selected item of the plurality of items will pass a second applied filter of the constraint filter and the recommendation filter, (iii) a cost of applying a first applied filter of the constraint filter and the recommendation filter to generate a single item, and (iv) a cost of applying the second applied filter of the constraint filter and the recommendation filter to the single item.

19. A system, comprising:

means for receiving a recommendation request from a user;

means for determining an order for applying a constraint filter and a recommendation filter to a plurality of items based on a cost of applying the constraint filter and the recommendation filter, wherein the constraint filter determines if each item satisfies a constraint and the recommendation filter assigns a predicted enjoyment value to each item; and

means for generating a recommendation list, comprising applying the constraint filter and the recommendation filter in the determined order.

20. The system of claim 19, wherein the cost is based at least on (i) a number of results required, (ii) a probability that a randomly selected item of the plurality of items will pass a second applied filter of the constraint filter and the recommendation filter, (iii) a cost of applying a first applied filter of the constraint filter and the recommendation filter to generate a single item, and (iv) a cost of applying the second applied filter of the constraint filter and the recommendation filter to the single item.

21. A method, comprising:

sending a recommendation request to a server; and

receiving a recommendation list generated by the server of a constraint filter and a recommendation filter in an order based on a cost of applying the constraint filter and the recommendation filter, wherein the constraint filter determines if each of a plurality of items satisfies a constraint and the recommendation filter assigns a predicted enjoyment value to each item.

22. The method of claim 21, wherein the cost is based at least on (i) a number of results required, (ii) a probability that a randomly selected item of a plurality of items will pass a second applied filter of the constraint filter and the recommendation filter, (iii) a cost of applying a first applied filter of the constraint filter and the recommendation filter to generate a single item, and (iv) a cost of applying the second applied filter of the constraint filter and the recommendation filter to the single item.

23. A method, comprising:

setting parameters for a constraint filter and a recommendation filter based on a received recommendation request;

determining respective costs of applying the constraint filter and the recommendation filter to items in a data processing system, the costs being associated with a probability a selected one of the items will pass both the constraint filter and the recommendation filter;

applying a first one of the constraint or recommendation filter having a lower one of the costs; and

applying a second one of the constraint or recommendation filter having a higher one of the costs.

24. The method of claim 23, further comprising requiring a predetermined number of requests as an aspect of the parameters.

25. The method of claim 24, further comprising basing the costs on a cost of applying the first one of the constraint filter and the recommendation filter to the first one of the items.

26. The method of claim 25, further comprising basing the costs on a cost of applying the second one of the constraint filter and the recommendation filter to the first one of the items.

27. The method of claim 23, wherein the determining of the costs is recursive.

28. The method of claim 23, wherein the recommendation filter assigns a predicted enjoyment value to each of the items, the predicted enjoyment value based at least one of previous purchases, previous comments and a particular rating.

29. The method of claim 23, further comprising associating generation interfaces and rejection interfaces with the constraint and recommendation filters.

30. A non-transitory computer readable medium having instructions stored thereon, execution of which, by a computing device, causes the computing device to perform operations comprising:

setting parameters for a constraint filter and a recommendation filter based on a received recommendation request;

determining respective costs of applying the constraint filter and the recommendation filter to items in a data processing system, the costs being associated with a probability a selected one of the items will pass both the constraint filter and the recommendation filter;

applying a first one of the constraint or recommendation filter having a lower one of the costs; and

applying a second one of the constraint or recommendation filter having a higher one of the costs.

31. A system, comprising:

means for setting parameters for a constraint filter and a recommendation filter based on a received recommendation request;

means for determining respective costs of applying the constraint filter and the recommendation filter to items in a data processing system, the costs being associated with a probability a selected one of the items will pass both the constraint filter and the recommendation filter;

means for applying a first one of the constraint or recommendation filter having a lower one of the costs; and

means for applying a second one of the constraint or recommendation filter having a higher one of the costs.