GENERATING METADATA FOR ASSOCIATION WITH A COLLECTION OF CONTENT ITEMS

• US 20100274782A1
• Filed: 10/02/2008
• Published: 10/28/2010
• Est. Priority Date: 10/08/2007
• Status: Active Grant

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Barbierei, Mauro, Weda, Johannes
• Granted Patent: US 8,321,456 B2
• Time in Patent Office: Days
• US Class Current: 707/723
• CPC Class Codes: G06F 16/48 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/635 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/68 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/683 using metadata automatically derived from the content
• CPC Class Codes: G06F 16/735 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/78 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/7847 using low-level visual features of the video content

Abstract

A method of automatically generating metadata for association with a collection of content items accessible to a system (1) for processing data included in the content items, includes obtaining sets of metadata associated with the content items individually, each set of metadata including at least one attribute value associated with the content item. At least one distribution of values of an attribute over the sets of metadata associated with the respective content items is analyzed. At least one attribute value is selected in dependence on the analysis. The selected attribute value(s) are processed to generate the metadata for association with the collection, and the generated metadata are made available to the system (1) for processing data included in the content items in connection with an identification of the collection of content items.

Claims

1. Method of automatically generating metadata for association with a collection of content items accessible to a system (1) for processing data included in the content items, including:

obtaining sets of metadata associated with the content items individually, each set of metadata including at least one attribute value associated with the content item,

analyzing at least one distribution of values of an attribute over the sets of metadata associated with the respective content items,

selecting at least one attribute value in dependence on the analysis,

processing the selected attribute value(s) to generate the metadata for association with the collection, and

making the generated metadata available to the system (1) for processing data included in the content items in connection with an identification of the collection of content items.

2. Method according to claim 1, including identifying the content items included in the collection using data maintained by a hierarchical file system.

3. Method according to claim 1, wherein selecting at least one attribute value in dependence on the analysis includes selecting at least one value of fewer than all attributes having values included in sets of metadata associated with content items in the collection individually.

4. Method according to claim 3, including selecting at least one value of each of a number of attributes, the number being based on at least one of:

user input, and

results of the analysis of at least one distribution of values of an attribute over the sets of metadata associated with the respective content items.

5. Method according to claim 1, wherein the step of selecting at least one attribute value includes selecting at least one attribute by comparative analysis of distributions of values over the sets of metadata associated with the respective content items individually of each of a plurality of attributes.

6. Method according to claim 5, wherein the comparative analysis includes, for each of a plurality of candidate attributes,

partitioning the collection of content items into disjoint sets according to similarity of values of that candidate attribute associated with the respective content items individually, and

ranking the attributes according to a factor based at least on a relative size of a largest set in the partition made for the candidate attribute, with respect to a size of the collection.

7. Method according to claim 5, wherein the comparative analysis includes ranking candidate attributes according to a factor based at least partly on priority values assigned to the candidate attributes.

8. Method according to claim 5, including selecting a number of values of a selected attribute for processing to generate the metadata for association with the collection, wherein the number is based on at least one of:

user input, and

a characteristic of the distribution of values of that attribute over the sets of metadata associated with the respective content items individually.

9. Method according to claim 1, wherein the step of processing the selected attribute value(s) to generate the metadata includes generating metadata including attribute value information corresponding to at least one selected attribute value and qualifying data based on an analysis of the distribution of values of the selected attribute over the sets of metadata associated with the respective content items individually.

10. Method according to claim 1, wherein the step of processing the selected attribute value(s) to generate the metadata includes synthesizing text data representing a descriptive phrase.

11. Method according to claim 1 including, upon selecting at least one attribute value in dependence on the analysis, repeating the analysis on sets of metadata associated with respective content items in the collection other than those including at least one attribute value corresponding to the attribute value(s) selected prior to repeating the analysis.

12. Method according to claim 1, wherein the step of processing the selected attribute value(s) to generate the metadata includes processing the selected attribute value(s) to generate metadata for comparison with metadata generated for a different collection of content items and, if the metadata generated for comparison differs by less than a certain minimum, selecting at least one further attribute value in dependence on an analysis of at least one distribution of values of an attribute over the sets of metadata associated with the respective content items, wherein at least the further selected attribute value(s) is/are processed to generate the metadata for association with the collection.

13. Method according to claim 1, wherein making the generated metadata available to the system 1 for processing data included in the content items includes storing the generated metadata in association with data maintained by a file system and identifying a group of files corresponding to the content items.

15. System according to claim 14, configured to carry out a method according to claim 1.

16. Computer program including a set of instructions capable, when incorporated in a machine-readable medium, of causing a system (1) having information processing capabilities to perform a method according to claim 1.

14. System for automatically generating metadata for association with a collection of content items, which system is configured:

to obtain sets of metadata associated with the content items individually, each set of metadata including at least one attribute value associated with the content item,

to analyze at least one distribution of values of an attribute over the sets of metadata associated with the respective content items,

to select at least one attribute value in dependence on the analysis, and

to process the selected attribute value(s) to generate the metadata for association with the collection.