Recommendation apparatus, recommendation method, and storage medium

• US 9,672,275 B2
• Filed: 03/25/2014
• Issued: 06/06/2017
• Est. Priority Date: 03/29/2013
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Shimizu, Tomoyuki
• Primary Examiner(s): Beausoliel, Jr., Robert
• Assistant Examiner(s): Rayyan, Susan F
• Application Number:
  US14/224,613
• Publication Number:
  US 20140297677A1
• Time in Patent Office: 1,169 Days
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence

Abstract

There is provided a technique capable of recommending more useful contents to a recommendation target user who is the current work user. A group of recommendation candidate contents is obtained from a similar work case similar to work by the recommendation target user. A relationship between an arbitrary user and the recommendation candidate contents is detected. Recommendation candidate contents having the detected relationship with the recommendation target user are recommended by priority from among the group of recommendation candidate contents.

Claims

1. A recommendation apparatus that recommends contents to a recommendation target user that is a current user of the recommendation apparatus, comprising:

a processor and memory;

a recommendation unit;

a content management unit; and

a relationship storage unit,

wherein a group of recommendation candidate contents is obtained from the content management unit, the group of recommendation candidate contents being candidates to be used by the recommendation target user,

wherein a relationship is detected between a user of a similar work case similar to work by the recommendation target user and used contents of the similar work case obtained from the content management unit,

wherein the used contents of the similar work case include contents other than the recommendation candidate contents, and the recommendation candidate contents include contents other than the used contents of the similar work case,

wherein the relationship between a user and contents is held in the relationship storage unit,

wherein the recommendation unit recommends recommendation candidate contents having the detected relationship with the recommendation target user by priority from among the group of recommendation candidate contents excluding the used contents of the similar work case,

wherein at least one of the units is implemented upon execution of instructions stored in the memory by the processor.

2. The recommendation apparatus according to claim 1, wherein the relationship stored in the relationship storage unit is written in an RDFS/OWL format.

3. The recommendation apparatus according to claim 1, wherein the relationship stored in the relationship storage unit is written in a graph format.

4. The recommendation apparatus according to claim 1, wherein the recommendation unit is configured to determine contents having the detected relationship with the recommendation target user by using a SPARQL language.

5. The recommendation apparatus according to claim 1, wherein the group of recommendation candidate contents is obtained from a similar work case including simultaneously used contents.

6. The recommendation apparatus according to claim 1, wherein the recommendation unit is configured to, if there is no recommendation candidate contents having the detected relationship with the recommendation target user, recommend recommendation candidate contents having a relationship of a superordinate concept with the recommendation target user.

7. The recommendation apparatus according to claim 1, wherein the recommendation unit comprises:

a similar case acquisition unit that obtains information about known similar work cases similar to certain user work from the content management unit;

a relationship attainment unit that refers to and attains a conceptual relationship between the similar work case user and the a group of simultaneously used contents in the similar work case obtained by the similar case acquisition unit;

a relationship content determination unit that performs a verification on the relationship to determine whether there are consistent contents in the group of recommendation contents;

a candidate content acquisition unit that obtains IDs of a group of recommendation candidate contents from the content management unit; and

a content evaluation unit that evaluates validity of recommended contents for the recommendation target user, based on a result of the relationship content determination unit,

wherein at least one of the units is implemented upon execution of instructions stored in the memory by the processor.

8. The recommendation apparatus according to claim 1, wherein the relationship is detected between a user of similar work case and the recommendation candidate contents obtained from the content management unit.

9. The recommendation apparatus according to claim 1, wherein contents used in a past similar work are excluded from the recommendation candidate contents.

10. A recommendation method performed in an apparatus that recommends contents to a recommendation target user that is a current user of the apparatus, the apparatus including units which are implemented upon execution of instructions stored in memory by a processor, the method comprising:

via a content management unit, obtaining a group of recommendation candidate contents, the group of recommendation candidate contents being candidates to be used by the recommendation target user,

detecting a relationship between a user of a similar work case similar to work by the recommendation target user and used contents of the similar work case obtained from the content management unit; and

via a recommendation unit, recommending recommendation candidate contents having the detected relationship with the recommendation target user by priority from among the group of recommendation candidate contents excluding the used contents of the similar work case,

wherein the used contents of the similar work case include contents other than the recommendation candidate contents, and the recommendation candidate contents include contents other than the used contents of the similar work case,

wherein the relationship between a user and contents is held in the relationship storage unit.

11. The recommendation method according to claim 10, further comprising:

via a similar case acquisition unit, obtaining information about known similar work cases similar to certain user work from the content management unit;

via a relationship attainment unit, referring to and attaining a conceptual relationship between the similar work case user and the a group of simultaneously used contents in the similar work case obtained by the similar case acquisition unit;

via a relationship content determination unit, performing a verification on the relationship to determine whether there are consistent contents in the group of recommendation contents;

via a candidate content acquisition unit, obtaining IDs of a group of recommendation candidate contents from the content management unit; and

via a content evaluation unit, evaluating validity of recommended contents for the recommendation target user, based on a result of the relationship content determination unit.

12. The medium according to claim 10, further comprising:

via a similar case acquisition unit, obtaining information about known similar work cases similar to certain user work from the content management unit;

via a relationship attainment unit, referring to and attaining a conceptual relationship between the similar work case user and the a group of simultaneously used contents in the similar work case obtained by the similar case acquisition unit;

via a relationship content determination unit, performing a verification on the relationship to determine whether there are consistent contents in the group of recommendation contents;

via a candidate content acquisition unit, obtaining IDs of a group of recommendation candidate contents from the content management unit; and

via a content evaluation unit, evaluating validity of recommended contents for the recommendation target user, based on a result of the relationship content determination unit.

13. A non-transitory computer-readable storage medium storing a program for causing a computer to perform a recommendation method comprising:

via a content management unit, obtaining a group of recommendation candidate contents, the group of recommendation candidate contents being candidates to be used by the recommendation target user;

detecting a relationship between a user of a similar work case similar to work by the recommendation target user and used contents of the similar work case obtained from the content management unit; and

via a recommendation unit, recommending recommendation candidate contents having the detected relationship with the recommendation target user by priority from among the group of recommendation candidate contents excluding the used contents of the similar work case,

wherein the used contents of the similar work case include contents other than the recommendation candidate contents, and the recommendation candidate contents include contents other than the used contents of the similar work case,

wherein the relationship between a user and contents is held in the relationship storage unit.