Content recommendation device and content recommendation method

• US 8,706,780 B2
• Filed: 09/16/2009
• Issued: 04/22/2014
• Est. Priority Date: 01/13/2009
• Status: Expired due to Fees

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Tokita, Eiji, Ozaki, Satoshi, Odaka, Kenji, Wakayama, Shirou, Fujimoto, Kensaku, Terashima, Yoshiki
• Primary Examiner(s): Mofiz, Apu
• Assistant Examiner(s): Bibbee, Jared
• Application Number:
  US12/560,761
• Publication Number:
  US 20100180203A1
• Time in Patent Office: 1,679 Days
• Field of Search: 707/736, 707/748, 707/749, 707/751, 707/723, 707/812
• US Class Current: 707/812
• CPC Class Codes: H04L 12/66 Arrangements for connecting between networks having differing types of switching systems, e.g. gateways

Abstract

A content item recommendation device includes; a storage which stores information of a plurality of content items describing identifiers and usage characteristics of the resources used in the content items; a first specifying unit which specifies first content items which use at least S number of resources in common with a specified content item; a second specifying unit which specifies second content items which use at least K number of resources in common with each of the first content items; a calculator which calculates relationship levels between the specified content item and each of the second content items; a selector which selects recommended content items from the second content items based on the relationship levels.

Claims

1. A content item recommendation device including a computer with a processor, comprising:

a storage configured to store a plurality of content item information for a plurality of content items each reproduced by using a plurality of resources, each content item information including identifiers of resources used in each content item and usage characteristics of the resources, and the content item information being obtained by analyzing script files for reproduction of the content item;

a receiver configured to receive an identifier of a specified content item;

a first specifying unit configured to specify first content items each of which uses at least S number (S is an integer of 1 or more) of resources in common with the specified content item, based on the content item information obtained by accessing the storage;

a second specifying unit configured to specify second content items each of which uses at least K number (K is an integer of 1 or more) of resources in common with at least one of the first content items, the second content items being specified from remaining content items except the specified content item and the first content items from among the content items, based on the content item information obtained by accessing the storage;

a relationship level calculator configured to calculate relationship levels between the specified content item and each of the second content items based on usage characteristics of target resources used in each of the second content items, the target resources being all resources used in common with said at least one of the first content items which uses said at least K number of resources in common;

a selector configured to select recommended content items from among the second content items based on the relationship levels of the second content items; and

an output unit configured to output a list of the recommended content items.

2. The device according to claim 1,

wherein the script files have a layer structure in which a script file of an upper layer calls a script file of a lower layer,

wherein the storage stores layer levels of the resources used in each of the content items as the use characteristics, and

wherein the relationship level calculator calculates the relationship levels with respect to the specified content item based on the layer levels of the target resources used in each of the second content items.

3. The device according to claim 2,

wherein point values are set for each layer level of the layer structure, and

wherein the relationship level calculator calculates the relationship levels in accordance with totals of the point values set for the layer levels of the target resources used in each of the second content items.

4. The device according to claim 1, wherein the relationship level calculator calculates the relationship levels with respect to the specified content item by further using usage characteristics of the target resources used in the at least one of the first content items.

5. The device according to claim 4,

wherein each of the script files defines a plurality of screen effects applied to the resources when being reproduced,

wherein the storage stores values of the screen effects applied to the resources used in each of the content items, as the use characteristics, and

wherein the relationship level calculator calculates the relationship levels with respect to the specified content item by whether each value of the screen effects applied to the target resources is matched or not between respective second content items and the one or plurality of the first content items.

6. The device according to claim 5,

wherein point values are set for a number or a ratio of matched effects, and

wherein the relationship level calculator calculates the relationship levels with respect to the specified content item in accordance with totals of the point values of the target resources used in each of the second content items.

7. The device according to claim 1, wherein the selector selects one or a plurality of the second content items which has the relationship level equal to or greater than a threshold value as the recommended content items.

8. The device according to claim 1, wherein the selector selects a predetermined number of the second content items in a descending order of the relationship levels as the recommended content items.

9. The device according to claim 1,

wherein the receiver receives the identifier of the specified content item from a content item reproducing device which is reproducing the specified content item, through a network, and

wherein the output unit transmits a list of the recommended content items to the content reproducing device through the network.

11. The device according to claim 1,

wherein the at least S is equal to 1.

12. The device according to claim 1, further comprising a content item information analyzer analyzing the script files upon receiving a registration request of the content item to thereby obtain the content item information including the identifiers and the usage characteristics of the resources.

13. The device according to claim 12, where the storage receives the content item information from the content item information analyzer and stores the content item information therein.

10. A content item recommendation method executed by a computer, comprising:

obtaining, by accessing a content item information storage, a plurality of content item information for a plurality of content items each reproduced by using a plurality of resources, each content item information including identifiers of resources used in each content item and usage characteristics of the resources, and the content item information being obtained by analyzing script files for reproduction of the content item;

specifying first content items each of which uses at least S number (S is an integer of 1 or more) of resources in common with the specified content item, based on the content item information storage;

specifying second content items each of which uses at least K number (K is an integer of 1 or more) of resources in common with at least one of the first content items, the second content items being specified from remaining content items except the specified content item and the first content items from among the content items, based on the content item information storage;

calculating relationship levels between the specified content item and each of the second content items based on usage characteristics of target resources used in each of the second content items, the target resources being all resources used in common with said at least one of the first content items which uses said at least K number of resources in common;

selecting recommended content items from among the second content items based on the relationship levels of the second content items; and

outputting a list of the recommended content items.