Item selecting apparatus, item selecting method and item selecting program

• US 9,740,982 B2
• Filed: 01/19/2015
• Issued: 08/22/2017
• Est. Priority Date: 04/28/2010
• Status: Active Grant

• Current Assignee: JVC Kenwood Corporation
• Original Assignee: JVC Kenwood Corporation
• Inventors: Shishido, Ichiro, Matsushita, Konosuke
• Primary Examiner(s): Cao, Phuong Thao
• Application Number:
  US14/599,845
• Publication Number:
  US 20150170041A1
• Time in Patent Office: 946 Days
• Field of Search: 707751
• CPC Class Codes: G06F 16/686 using information manually generated, e.g. tags, keywords, comments, title or artist information, time, location or usage information, user ratings
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

In an item selecting apparatus performing a selection of an item to be recommended for each user, it is performed to calculate, with respect to each usage registration of an item by a user, an elapsed value as a difference between a time point of creating the item or staring providing of the item and a predetermined time point, acquire a usage characteristics of each user based on the elapsed value and calculate a freshness value representing a degree of freshness about each item. Further, using correspondence rules of different characteristics corresponding to the usage characteristics, it is performed to calculate a novelty index by applying the freshness value of each item on the correspondence rule corresponding to the usage characteristics of each user, calculate a priority of the item for each user, based on the novelty index and performing a selection of the item.

Claims

1. An item selecting apparatus comprising:

a usage characteristics information calculator configured to;

calculate an elapsed value as a difference between a time point of item time information representing a time point of creating an item or a time point of starting providing of the item and a given time point; and

calculate usage characteristics information of a use subject with respect to item usage conditions based on a distribution of elapsed values calculated for the use subject; and

an item selector configured to;

select a correspondence rule in accordance with the usage characteristics information from a plurality of correspondence rules with different characteristics;

calculate a novelty index for each item using the correspondence rule and the item time information;

calculate a popularity index as a numerical value associated with the number of usages for each item or the number of use subjects for each item;

calculate a priority that takes a higher value as the novelty index takes a higher value and the popularity index takes a higher value; and

select a recommendation item for the use subject based on the priority;

wherein the usage characteristics information calculator and the item selector are hardware or a combination of hardware and software.

2. The item selecting apparatus according to claim 1, wherein

the usage characteristics information calculator calculates the usage characteristics information for the use subject based on elapsed values calculated as a respective difference between:

a time point of item time information of each item included in a usage history representing items used by the use subject; and

a given time point.

3. The item selecting apparatus according to claim 1, wherein

the item selector calculates the novelty index by applying a value, which is based on the item time information, on the correspondence rule.

4. The item selecting apparatus according to claim 1, wherein

the correspondence rule is:

a function representing a relationship between a value, which is based on the item time information, and the novelty index;

an input-output correspondence rule;

or an input-output correspondence table.

5. The item selecting apparatus according to claim 1, wherein

the given time point is a time point of usage of an item, or a time point of reception of usage information by a receiver.

6. The item selecting apparatus according to claim 1, wherein

the item selector selects an item which belongs to a category satisfying a condition.

7. The item selecting apparatus according to claim 1, wherein

the item selector performs a selection of a recommendation item based on an evaluated value representing an evaluation level against an item.

8. The item selecting apparatus according to claim 1, wherein

the usage characteristics information calculator uses a representative value, which is based on the distribution of the elapsed values, as the usage characteristics information.

9. The item selecting apparatus according to claim 1, wherein

the usage characteristics information calculator uses a degree of dispersion of the elapsed values as the usage characteristics information.

10. The item selecting apparatus according to claim 1, wherein

the usage characteristics information calculator uses the number of elapsed values as the usage characteristics information.

11. The item selecting apparatus according to claim 1, further comprising:

a similarity calculator configured to calculate a degree of similarity between any two use subjects,

wherein the item selector performs a selection of a recommendation item by using a similarity index based on the degree of similarity, and

wherein the similarity calculator is hardware or a combination of hardware and software.

12. The item selecting apparatus according to claim 1, further comprising:

a similarity calculator configured to calculate a degree of similarity between any two use subjects,

wherein the item selector extracts other use subjects whose each degree of similarity to the use subject is higher than a given value or other use subjects whose number is not in excess of a predetermined number, in descending order of the degree of similarity to the use subject; and

performs a selection of a recommendation item for the use subject from a collection of items used by extracted use subjects, and

wherein the similarity calculator is hardware or a combination of hardware and software.

13. The item selecting apparatus according to claim 1, further comprising:

a relevance calculator configured to calculate a relevance between any two use subjects,

wherein the item selector performs a selection of a recommendation item based on the relevance, and

wherein the relevance calculator is hardware or a combination of hardware and software.

14. The item selecting apparatus according to claim 1, further comprising:

a relevance calculator configured to calculate a relevance between any two use subjects,

wherein the item selector extracts other use subjects whose each relevance to the use subject is higher than a given value or other use subjects whose number is not in excess of a predetermined number, in descending order of the relevance to the use subject; and

performs a selection of a recommendation item for the use subject from a collection of items used by extracted use subjects, and

wherein the relevance calculator is hardware or a combination of hardware and software.

15. An item selecting method performed with a processor or a computing system for selecting an item, comprising the steps of:

calculating an elapsed value as a difference between a time point of item time information representing a time point of creating an item or a time point of starting providing of the item and a given time point;

calculating usage characteristics information of a use subject with respect to item usage conditions based on a distribution of elapsed values calculated for the use subject;

selecting a correspondence rule in accordance with the usage characteristics information from a plurality of correspondence rules with different characteristics;

calculating a novelty index for each item using the correspondence rule and the item time information;

calculating a popularity index as a numerical value associated with the number of usages for each item or the number of use subjects for each item;

calculating a priority that takes a higher value as the novelty index takes a higher value and the popularity index takes a higher value; and

selecting a recommendation item for the use subject based on the priority.

16. A non-transitory computer readable medium storing an item selecting program for causing a computer to execute the steps of:

calculating an elapsed value as a difference between a time point of item time information representing a time point of creating an item or a time point of starting providing of the item and a given time point;

calculating usage characteristics information of a use subject with respect to item usage conditions based on a distribution of elapsed values calculated for the use subject;

selecting a correspondence rule in accordance with the usage characteristics information from a plurality of correspondence rules with different characteristics;

calculating a novelty index for each item using the correspondence rule and the item time information;

calculating a popularity index as a numerical value associated with the number of usages for each item or the number of use subjects for each item;

calculating a priority that takes a higher value as the novelty index takes a higher value and the popularity index takes a higher value; and

selecting a recommendation item for the use subject based on the priority.