ITEM SELECTING APPARATUS, ITEM SELECTING METHOD AND ITEM SELECTING PROGRAM

• US 20130046766A1
• Filed: 04/21/2011
• Published: 02/21/2013
• Est. Priority Date: 04/28/2010
• Status: Active Grant

• Current Assignee: Rakuten Group, Inc.
• Original Assignee: JVC Kenwood Corporation
• Inventors: Shishido, Ichiro, Matsushita, Konosuke
• Granted Patent: US 8,972,419 B2
• Time in Patent Office: Days
• US Class Current: 707/741
• 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 (S420), calculate a priority of the item for each user, based on the novelty index (S430) and performing a selection of the item (S440).

Claims

1. An item selecting apparatus comprising:

an item property storage unit configured to store an item identifier of each item and item time information representing a time point of creating the item or a time point of starting providing of the item, in an associated manner;

a receiving unit configured to receive usage information from another apparatus, the usage information including a use subject identifier for identifying either a user or a terminal unit that the user has used and an item identifier for identifying an item used by the user;

a usage history storage unit configured to store a usage history which associates the use subject identifier with the item identifier contained in the usage information received by the receiving unit;

a usage characteristics information calculating unit configured to calculate an elapsed value as a difference between the item time information corresponding to the item identifier contained in the usage history stored in the usage history storage unit and a given time point and calculate usage characteristics information with respect to each use subject identifier contained in the usage history stored in the usage history storage unit, based on a distribution of the elapsed values corresponding to the use subject identifier; and

an item selecting unit configured to;

calculate, with respect to the item identifier stored in the item property storage unit, a freshness value representing a degree of freshness of the item related to the item identifier, based on the item time information corresponding to the item identifier;

calculate a novelty index by using correspondence rules with different characteristics corresponding to the usage characteristics information and further applying the freshness value of them on the correspondence rule corresponding to the usage characteristics information of each use subject identifier;

calculate a priority of the item each use subject identifier, based on the novelty index; and

perform a selection of the item each use subject identifier, based on the priority.

3. The item selecting apparatus of claim 1, wherein

the usage history storage unit stores either a usage history associating three information contained in the usage information received by the receiving unit with each other, the three information comprising the use subject identifier, the item identifier and usage time information representing a time point when the item was used, or another usage history associating two information contained in the usage information received by the receiving unit and usage time information representing a time point when the receiving unit received the usage information with each other, the two information comprising the use subject identifier and the item identifier, and

the usage characteristics information calculating unit adopts the usage time information contained in the usage history as the given time point, in calculating the elapsed values about the usage history stored in the usage history storage unit.

7. The item selecting apparatus of claim 1, wherein

in storing the item identifier of the item and the item time information in the associated manner, the item property storage unit stores category information representing a category where the item belongs to, in a further-associated manner with the item identifier and the item time information, and

the item selecting unit selects an item related to the item identifier corresponding to the category information satisfying a predetermined condition, in performing a selection of the item with respect to each use subject identifier based on the priority.

8. The item selecting apparatus of claim 1, wherein the item selecting unit calculates a popularity index as a numerical value related to either the number of use times with respect to each item or the number of use users with respect to each item, based on the usage history stored in the usage history storage unit, and performs calculations of the priority by using the popularity index.

9. The item selecting apparatus of claim 1, wherein

the usage information received from the other apparatus contains an evaluated value representing an evaluation level of a user related to the usage information against an item related to the usage information,

the usage history stored based on the usage information received by the receiving unit is associated with the evaluated value contained in the usage information, and

the item selecting unit calculates a popularity index by using a value obtained by adding the evaluated value with respect to each item, based on the usage history stored in the usage history storage unit, and performs calculations of the priority by using the popularity index.

10. The item selecting apparatus of claim 1, wherein the usage characteristics information calculating unit calculates a representative value based on the distribution of the elapsed values, as a constituent of the usage characteristic information, in calculating, with respect to each use subject identifier, the usage characteristics information based on the distribution of elapsed values corresponding to the use subject identifier.

11. The item selecting apparatus of claim 1, wherein the usage characteristics information calculating unit calculates a degree of dispersion of the elapsed values, as a constituent of the usage characteristic information, in calculating, with respect to each use subject identifier, the usage characteristics information based on the distribution of elapsed values corresponding to the use subject identifier.

12. The item selecting apparatus of claim 1, wherein the usage characteristics information calculating unit calculates the number of elapsed values, as a constituent of the usage characteristic information, in calculating, with respect to each use subject identifier, the usage characteristics information based on the distribution of elapsed values corresponding to the use subject identifier.

13. The item selecting apparatus of claim 1, further comprising a similarity calculating unit configured to calculate a degree of similarity between any two use subject identifiers, based on the usage history stored in the usage history storage unit, wherein the item selecting unit performs calculations of the priority by using a similarity index based on the degree of similarity.

14. The item selecting apparatus of claim 1, further comprising a similarity calculating unit configured to calculate a degree of similarity between any two use subject identifiers, based on the usage history stored in the usage history storage unit, wherein

in calculating the priority of an item with respect to each use subject identifier, the item selecting unit:

extracts, with respect to each use subject identifier, the other use subject identifiers whose each degree of similarity to the each use subject identifier is higher than a given value or the other use subject identifiers whose number is not in excess of a predetermined number, in descending order of the degree of similarity to the each use subject identifier;

extracts the item identifiers associated with the extracted use subject identifiers from the usage history stored in the usage history storage unit; and

performs calculations of the priority while targeting at a group of items related to the extracted item identifiers.

15. The item selecting apparatus of claim 1, further comprising:

a user property storage unit configured to store the use subject identifier for identifying the user or the terminal unit that the user has used and property information of the user in an associated manner; and

a relevance calculating unit configured to calculate a relevance between any two use subject identifiers based on the user property information stored in the user property storage unit, wherein

the item selecting unit performs calculations of the priority by using a relevance index based on the relevance.

16. The item selecting apparatus of claim 1, further comprising:

a user property storage unit configured to store the use subject identifier for identifying the user or the terminal unit that the user has used and property information of the user in an associated manner; and

a relevance calculating unit configured to calculate a relevance between any two use subject identifiers based on the user property information stored in the user property storage unit, wherein

in calculating the priority of an item with respect to each use subject identifier, the item selecting unit;

extracts, with respect to each use subject identifier, the other use subject identifiers whose each relevance to the each use subject identifier is higher than a given value or the other use subject identifiers whose number is not in excess of a predetermined number, in descending order of the relevance to the each use subject identifier;

extracts the item identifiers associated with the extracted use subject identifiers from the usage history stored in the usage history storage unit; and

performs calculations of the priority while targeting at a group of items related to the extracted item identifiers.

2. An item selecting apparatus comprising:

a receiving unit configured to receive usage information from another apparatus, the usage information including at least an use subject identifier for identifying either a user or a terminal unit that the user has used and an item identifier for identifying an item used by the user;

a usage history storage unit configured to store either a usage history associating three information contained in the usage information received by the receiving unit with each other, the three information comprising the use subject identifier, the item identifier and usage time information representing a time point when the item was used, or another usage history associating two information contained in the usage information received by the receiving unit and usage time information representing a time point when the receiving unit received the usage information with each other, the two information comprising the use subject identifier and the item identifier;

a usage characteristics information calculating unit configured to calculate an elapsed value as a difference between the usage time information contained in the usage history stored in the usage history storage unit and a given time point and calculate usage characteristics information, with respect to each use subject identifier contained in the usage history stored in the usage history storage unit, based on a distribution of the elapsed values corresponding to the use subject identifier; and

an item selecting unit configured to;

calculate, with respect to each item identifier stored in the usage history storage unit, a freshness value representing a degree of freshness of the item related to the item identifier, based on the usage time information corresponding to the item identifier;

calculating a novelty index by using correspondence rules with different characteristics corresponding to the usage characteristics information and further applying the freshness value on the correspondence rule of them corresponding to the usage characteristics information of each use subject identifier;

calculate a priority of the item each use subject identifier, based on the novelty index; and

perform a selection of the item each use subject identifier, based on the priority.

4. The item selecting apparatus of claim 2, wherein the usage characteristics information calculating unit calculates estimated item time information as an estimate value for a time point of creating the item or starting providing of the item related to the item identifier by using a distribution of the usage time information of respective usage histories containing the item identifier of the usage history, and adopts the estimated item time information as the given time point, in calculating the elapsed value about the usage history stored in the usage history storage unit:

5. The item selecting apparatus of claim 2, wherein the usage characteristics information calculating unit adopts, as the given time point, the usage time information contained in the other usage history which contains an use subject identifier identical to the use subject identifier contained in the usage history and which temporally-neighbors the usage history, in calculating the elapsed value about the usage history stored in the usage history storage unit.

6. The item selecting apparatus of claim 5, further comprising an item property storage unit configured to store an item identifier of each item and item time information representing a time point of creating the item or a time point of starting providing of the item, in an associated manner, wherein

the item selecting unit calculates the freshness value based on the item time information stored in the item property storage unit, in place of the usage time information.

21. The item selecting apparatus of claim 2, wherein

in storing the item identifier of the item and the item time information in the associated manner, the item property storage unit stores category information representing a category where the item belongs to, in a further-associated manner with the item identifier and the item time information, and

the item selecting unit selects an item related to the item identifier corresponding to the category information satisfying a predetermined condition, in performing a selection of the item with respect to each use subject identifier based on the priority.

22. The item selecting apparatus of claim 2, wherein the item selecting unit calculates a popularity index as a numerical value related to either the number of use times with respect to each item or the number of use users with respect to each item, based on the usage history stored in the usage history storage unit, and performs calculations of the priority by using the popularity index.

23. The item selecting apparatus of claim 2, wherein

the usage information received from the other apparatus contains an evaluated value representing an evaluation level of a user related to the usage information against an item related to the usage information,

the usage history stored based on the usage information received by the receiving unit is associated with the evaluated value contained in the usage information, and

the item selecting unit calculates a popularity index by using a value obtained by adding the evaluated value with respect to each item, based on the usage history stored in the usage history storage unit, and performs calculations of the priority by using the popularity index.

24. The item selecting apparatus of claim 2, wherein the usage characteristics information calculating unit calculates a representative value based on the distribution of the elapsed values, as a constituent of the usage characteristic information, in calculating, with respect to each use subject identifier, the usage characteristics information based on the distribution of elapsed values corresponding to the use subject identifier.

25. The item selecting apparatus of claim 2, wherein the usage characteristics information calculating unit calculates a degree of dispersion of the elapsed values, as a constituent of the usage characteristic information, in calculating, with respect to each use subject identifier, the usage characteristics information based on the distribution of elapsed values corresponding to the use subject identifier.

26. The item selecting apparatus of claim 2, wherein the usage characteristics information calculating unit calculates the number of elapsed values, as a constituent of the usage characteristic information, in calculating, with respect to each use subject identifier, the usage characteristics information based on the distribution of elapsed values corresponding to the use subject identifier.

27. The item selecting apparatus of claim 2, further comprising a similarity calculating unit configured to calculate a degree of similarity between any two use subject identifiers, based on the usage history stored in the usage history storage unit, wherein the item selecting unit performs calculations of the priority by using a similarity index based on the degree of similarity.

28. The item selecting apparatus of claim 2, further comprising a similarity calculating unit configured to calculate a degree of similarity between any two use subject identifiers, based on the usage history stored in the usage history storage unit, wherein

in calculating the priority of an item with respect to each use subject identifier, the item selecting unit:

extracts, with respect to each use subject identifier, the other use subject identifiers whose each degree of similarity to the each use subject identifier is higher than a given value or the other use subject identifiers whose number is not in excess of a predetermined number, in descending order of the degree of similarity to the each use subject identifier;

extracts the item identifiers associated with the extracted use subject identifiers from the usage history stored in the usage history storage unit; and

performs calculations of the priority while targeting at a group of items related to the extracted item identifiers.

29. The item selecting apparatus of claim 2, further comprising:

a user property storage unit configured to store the use subject identifier for identifying the user or the terminal unit that the user has used and property information of the user in an associated manner; and

a relevance calculating unit configured to calculate a relevance between any two use subject identifiers based on the user property information stored in the user property storage unit, wherein

the item selecting unit performs calculations of the priority by using a relevance index based on the relevance.

30. The item selecting apparatus of claim 2, further comprising:

a user property storage unit configured to store the use subject identifier for identifying the user or the terminal unit that the user has used and property information of the user in an associated manner; and

a relevance calculating unit configured to calculate a relevance between any two use subject identifiers based on the user property information stored in the user property storage unit, wherein

in calculating the priority of an item with respect to each use subject identifier, the item selecting unit;

extracts, with respect to each use subject identifier, the other use subject identifiers whose each relevance to the each use subject identifier is higher than a given value or the other use subject identifiers whose number is not in excess of a predetermined number, in descending order of the relevance to the each use subject identifier;

extracts the item identifiers associated with the extracted use subject identifiers from the usage history stored in the usage history storage unit; and

performs calculations of the priority while targeting at a group of items related to the extracted item identifiers.

17. An item selecting method comprising the steps of:

storing an item identifier of each item and item time information representing a time point of creating the item or a time point of starting providing of the item in an item property storage unit, in an associated manner;

receiving usage information including a use subject identifier for identifying either a user or a terminal unit that the user has used and an item identifier for identifying an item used by the user, from another apparatus;

storing a usage history which associates the use subject identifier with the item identifier contained in the usage information received by the receiving unit, in a usage history storage unit;

calculating an elapsed value as a difference between the item time information corresponding to the item identifier contained in the usage history stored in the usage history storage unit and a given time point;

calculating usage characteristics information with respect to each use subject identifier contained in the usage history stored in the usage history storage unit, based on a distribution of the elapsed values corresponding to the use subject identifier;

calculating a freshness value representing a degree of freshness of the item related to the item identifier, based on the item time information corresponding to the item identifier, with respect to each item identifier stored in the item property storage unit;

calculating a novelty index by using correspondence rules with different characteristics corresponding to the usage characteristics information and applying the freshness value on the correspondence rule of them corresponding to the usage characteristics information of each use subject identifier;

calculating a priority of the item each use subject identifier, based on the novelty index; and

performing a selection of the item each use subject identifier, based on the priority.

18. An item selecting method comprising the steps of:

receiving usage information including at least an use subject identifier for identifying either a user or a terminal unit that the user has used and an item identifier for identifying an item used by the user, from another apparatus;

storing either a usage history associating three information contained in the usage information received by the receiving unit with each other, the three information comprising the use subject identifier, the item identifier and usage time information representing a time point when the item was used, or another usage history associating two information contained in the usage information received by the receiving unit and usage time information representing a time point when the receiving unit received the usage information with each other, the two information comprising the use subject identifier and the item identifier;

calculating an elapsed value as a difference between the usage time information contained in the usage history stored in the usage history storage unit and a given time point;

calculating usage characteristics information, with respect to each use subject identifier contained in the usage history stored in the usage history storage unit, based on a distribution of the elapsed values corresponding to the use subject identifier;

calculating a freshness value representing a degree of freshness of the item related to the item identifier, based on the usage time information corresponding to the item identifier, with respect to each item identifier stored in the item property storage unit;

calculating a novelty index by using correspondence rules with different characteristics corresponding to the usage characteristics information and applying the freshness value on the correspondence rule of them corresponding to the usage characteristics information of each use subject identifier;

calculating a priority of the item each use subject identifier, based on the novelty index; and

performing a selection of the item each use subject identifier, based on the priority.

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

storing an item identifier of each item and item time information representing a time point of creating the item or a time point of starting providing of the item in an item property storage unit, in an associated manner;

receiving usage information including a use subject identifier for identifying either a user or a terminal unit that the user has used and an item identifier for identifying an item used by the user, from another apparatus;

storing a usage history which associates the use subject identifier with the item identifier contained in the usage information received by the receiving unit, in a usage history storage unit;

calculating an elapsed value as a difference between the item time information corresponding to the item identifier contained in the usage history stored in the usage history storage unit and a given time point;

calculating usage characteristics information with respect to each use subject identifier contained in the usage history stored in the usage history storage unit, based on a distribution of the elapsed values corresponding to the use subject identifier;

calculating a freshness value representing a degree of freshness of the item related to the item identifier, based on the item time information corresponding to the item identifier, with respect to each item identifier stored in the item property storage unit;

calculating a novelty index by using correspondence rules with different characteristics corresponding to the usage characteristics information and applying the freshness value on the correspondence rule of them corresponding to the usage characteristics information of each use subject identifier;

calculating a priority of the item each use subject identifier, based on the novelty index; and

performing a selection of the item each use subject identifier, based on the priority.

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

receiving usage information including at least an use subject identifier for identifying either a user or a terminal unit that the user has used and an item identifier for identifying an item used by the user, from another apparatus;

storing either a usage history associating three information contained in the usage information received by the receiving unit with each other, the three information comprising the use subject identifier, the item identifier and usage time information representing a time point when the item was used, or another usage history associating two information contained in the usage information received by the receiving unit and usage time information representing a time point when the receiving unit received the usage information with each other, the two information comprising the use subject identifier and the item identifier;

calculating an elapsed value as a difference between the usage time information contained in the usage history stored in the usage history storage unit and a given time point;

calculating usage characteristics information, with respect to each use subject identifier contained in the usage history stored in the usage history storage unit, based on a distribution of the elapsed values corresponding to the use subject identifier;

calculating a freshness value representing a degree of freshness of the item related to the item identifier, based on the usage time information corresponding to the item identifier, with respect to each item identifier stored in the item property storage unit;

calculating a novelty index by using correspondence rules with different characteristics corresponding to the usage characteristics information and applying the freshness value on the correspondence rule of them corresponding to the usage characteristics information of each use subject identifier;

calculating a priority of the item each use subject identifier, based on the novelty index; and