Media agent

• US 7,493,312 B2
• Filed: 05/24/2004
• Issued: 02/17/2009
• Est. Priority Date: 11/30/2001
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Liu, Wen-Yin, Chen, Zheng, Zhang, Hong-Jiang
• Primary Examiner(s): TRUONG, CAM Y T
• Application Number:
  US10/853,375
• Publication Number:
  US 20040220925A1
• Time in Patent Office: 1,730 Days
• Field of Search: 707 3- 5, 707/7, 707/10
• US Class Current: 1/1
• CPC Class Codes: G06F 16/40 of multimedia data, e.g. slideshows comprising image and additional audio data retrieval of still image data G06F16/50; retrieval of audio data G06F16/60; retrieval of video data G06F16/70
• 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: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99937 Sorting
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

Systems and methods for a media agent are described. In one aspect, user access of a media content source is detected. Responsive to this detection, a piece of media content and associated text is collected from the media content source. Semantic text features are extracted from the associated text and the piece of media content. The semantic text features are indexed into a media database.

Claims

1. A computer-implemented method for indexing a database, the method comprising:

detecting user access of media content source by using a computer, the media content source is stored in a computer readable storage medium;

determining media content use preferences based on user actions, wherein the media content use preferences comprise a plurality of user preference models, each user preference model comprising semantically similar keywords that correspond to the user actions, wherein the media content use preferences are further based on keywords extracted from information corresponding to the user actions;

automatically collecting, from the media content source, media content referenced by a uniform resource locator (URL) and the URL, wherein the detecting, the determining and the collecting are performed responsive to detecting an idle state;

extracting semantic text features from, the media content and the URL, wherein the extracting the semantic text features is determined according to a set of extraction rules, wherein the set of extraction rules comprises at least one rule specifying surrounding text be within a threshold semantic distance from a piece of the media content;

determining that the media content is interest to a user based at least in part on semantic similarity between the media content use preferences and the semantic text features, and wherein the media content is determined to be interest to the user if there is semantic similarity between the media content and at least one of the user preference models;

responsive to determining that the media content is interest to the user, indexing a media database with the extracted semantic text features;

monitoring the user actions to identify patterns of media content use from the user actions;

modifying indication of relevancy with respect to ones of pieces of the media content based on the patterns of the media content use;

performing an analysis of a text to determine that the user desires to access the media content, wherein the analysis is based at least in part on patterns of previous media content use;

in response to performing the analysis, generating search criteria based on linguistic features of the text,

identifying, based at least in part on the search criteria, one or more media files that are semantically related to the text from the media database; and

presenting information corresponding to the one or more media files to the user.

2. The method as recited in claim 1, wherein the set of extraction rules comprise at least one rule specifying that the URL be parsed for words contained in a dictionary.

3. A computer-readable storage medium embodied computer-executable instructions that are executed by a computer for performing steps of:

determining media content use preferences based on user actions, wherein the media content use preferences comprise a plurality of user preference models, each user preference model comprising semantically similar keywords that correspond to the user actions;

detecting user access of a media content source;

in response the detecting the user access, automatically collecting, from the media content source, a piece of media content having a graphical presentation that includes surrounding text, and the surrounding text, wherein the determining, the detecting, and the collecting are performed responsive to detecting an idle state;

extracting semantic text features from, the surrounding text and the piece of media content, wherein the extracting the semantic text features is determined according to a set of extraction rules, wherein the set of extraction rules comprise at least one rule specifying the surrounding text be within a threshold graphical distance from the piece of media content,

wherein at least a portion of the semantic text features include an indication of relevancy with respect to ones of the pieces of media content;

monitoring the user actions to identify patterns of media content use from the user actions;

for each semantic text feature in the at least a portion, modifying the indication of relevancy with respect to ones of the pieces of media content based on the patterns of the media content use,

determining that the media content is interest to a user based at least in part on semantic similarity between the media content use preferences and the semantic text features, and wherein media content is determined to be interest to the user if there is semantic similarity between the media content and at least one of the user preference models;

responsive to determining that the media content is interest to the user, indexing a media database with the extracted semantic text features;

performing an analysis of a text to determine that the user desires to access the media content;

in response to performing the analysis, generating search criteria based on linguistic features of the text, identifying, based at least in part on the search criteria, one or more media files that are semantically related to the text from the media database; and

presenting information corresponding to the one or more media files to the user.

4. The computer-readable storage medium as recited in claim 3, wherein the media content source is an e-mail message, wherein the media content is an attachment to the e-mail message or a link, and wherein the text is a body of the e-mail message.

5. The computer-readable storage medium as recited in claim 3, wherein a media content source comprises a word processing document, wherein the media content is inserted media content, and wherein the text is document text.

6. A computing device for indexing a database, comprising:

a processor;

a memory coupled to the processor, the memory comprising computer-executable instructions, the computer-executable instructions are executed by the processor for performing steps of;

detecting user access of media content source by using a computer;

determining media content use preferences based on user actions, wherein the media content use preferences comprise a plurality of user preference models, each user preference model comprising semantically similar keywords that correspond to the user actions, wherein the media content use preferences are further based on keywords extracted from information corresponding to the user actions;

automatically collecting, from the media content source, media content referenced by a uniform resource locator (URL) and the URL, wherein the detecting, the determining and the collecting are performed responsive to detecting an idle state;

extracting semantic text features from, the media content and the URL, wherein the extracting the semantic text features is determined according to a set of extraction rules, wherein the set of extraction rules comprises at least one rule specifying surrounding text be within a threshold semantic distance from a piece of the media content;

determining that the media content is interest to a user based at least in part on semantic similarity between the media content use preferences and the semantic text features, and wherein the media content is determined to be interest to the user if there is semantic similarity between the media content and at least one of the user preference models;

responsive to determining that the media content is interest to the user, indexing a media database with the extracted semantic text features;

monitoring the user actions to identify patterns of media content use from the user actions;

modifying indication of relevancy with respect to ones of pieces of the media content based on the patterns of the media content use;

performing an analysis of a text to determine that the user desires to access the media content, wherein the analysis is based at least in part on patterns of previous media content use;

in response to performing the analysis, generating search criteria based on linguistic features of the text,

identifying, based at least in part on the search criteria, one or more media files that are semantically related to the text from the media database; and

presenting information corresponding to the one or more media files to the user.

7. A computing device for indexing a database, comprising:

a processor;

a memory coupled to the processor, the memory comprising computer-executable instructions, the computer-executable instructions are executed by the processor for performing steps of;

determining media content use preferences based on user actions, wherein the media content use preferences comprise a plurality of user preference models, each user preference model comprising semantically similar keywords that correspond to the user actions;

detecting user access of a media content source;

in response the detecting the user access, automatically collecting, from the media content source, a piece of media content having a graphical presentation that includes surrounding text, and the surrounding text, wherein the determining, the detecting, and the collecting are performed responsive to detecting an idle state;

extracting semantic text features from, the surrounding text and the piece of media content, wherein the extracting the semantic text features is determined according to a set of extraction rules, wherein the set of extraction rules comprise at least one rule specifying the surrounding text be within a threshold graphical distance from the piece of media content,

wherein at least a portion of the semantic text features include an indication of relevancy with respect to ones of the pieces of media content;

monitoring the user actions to identify patterns of media content use from the user actions;

for each semantic text feature in the at least a portion, modifying the indication of relevancy with respect to ones of the pieces of media content based on the patterns of the media content use,

determining that the media content is interest to a user based at least in part on semantic similarity between the media content use preferences and the semantic text features, and wherein media content is determined to be interest to the user if there is semantic similarity between the media content and at least one of the user preference models;

responsive to determining that the media content is interest to the user, indexing a media database with the extracted semantic text features;

performing an analysis of a text to determine that the user desires to access the media content;

in response to performing the analysis, generating search criteria based on linguistic features of the text, identifying, based at least in part on the search criteria, one or more media files that are semantically related to the text from the media database; and

presenting information corresponding to the one or more media files to the user.