Information Access With Usage-Driven Metadata Feedback

• US 20080097985A1
• Filed: 10/13/2006
• Published: 04/24/2008
• Est. Priority Date: 10/13/2005
• Status: Active Grant

• Current Assignee: Microsoft International Holdings BV (Microsoft Corporation)
• Original Assignee: Fast Search & Transfer ASA (Microsoft Corporation)
• Inventors: Moen, Christian, Olstad, Bjorn
• Granted Patent: US 8,135,669 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/907 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually

Abstract

In a method for displaying and capturing metadata of documents within result presentations in information access or search systems, a metadata server is used for storing a metadata associated with any searchable document and the end users are given the opportunity to view and edit metadata associated with documents returned from the metadata server which is capable of automatically creating metadata objects associated with any combination of document query and document position in a result set for a given query A search engine capable of implementing the method comprises a metadata server as part of or connected with its core search engine.

Claims

1. A method for displaying, capturing and using document metadata within result presentations in information access systems, or in information search systems, wherein the method comprises steps for

applying a query to a set of documents,

identifying matching documents, and

producing a result page with at least one of the most relevant documents,

wherein the method includes

using a metadata server capable of storing metadata associated with any searchable document, and

providing means within the result page for end users of the information access or search system to view and edit metadata associated with the returned documents from the metadata server, and

wherein the method is characterized by using the metadata server for automatically creating metadata objects associated with any combination of document, query and document position in the result set for the query.

2. A method according to claim 1,

characterized by the metadata server and viewing/editing of metadata providing means for associating metadata with a document subsection.

4. A method according to claim 1,

characterized by said metadata server uses blogs and/or wikis to create, view, edit and store metadata objects.

5. A method according to claim 1,

characterized by the metadata objects managed by the metadata server including at least one of free text, audio clips, video clips, images, predefined entries, emotion icons, values to predefined attributes, user defined attributes with associated values, links to associated documents or links to associated people, and preferably are then RDF or OWL used as mechanisms to encode the said metadata.

6. A method according to claim 1,

characterized by using access rights to control viewing and editing capabilities for the said metadata, whereby preferably document access rights in LDAP or Active Directory are used to control the said viewing and editing capabilities for the associated metadata, and wherein preferably user-defined groups can share viewing and editing of document metadata in a secure manner within communities.

7. A method according to claim 1,

characterized by means for improving the information access quality, said means comprising analyzing the metadata and deriving attributes relevant for search, and

making the information access system responsive to said relevant attributes, said analyzing of metadata preferably including exposing free text and searchable attributes for further content refinement and indexing in the information access system or search system.

8. A method according to claim 7,

characterized by making a ranking of relevant documents responsive to the metadata associated with the documents, and preferably boosting the relevance of the associate documents with one or more of an amount of metadata viewing and editing activity, a number of recent updates in the metadata, a number of explicit positive endorsements, or a sentiment analysis of the free text of the metadata to derive an automatic sentiment score.

9. A method according to claim 7,

characterized by modifying the information access system or search to be responsive to said derived attributes from associated metadata objects.

10. A method according to claim 1,

characterized by further comprising steps for detecting and removing unwanted content in the captured metadata, preferably presenting a so-called “

human challenge”

to avoid computer-generated metadata, and/or detecting and removing preferably offensive metadata entries.

11. A method according to claim 5 or claim 8,

characterized by identifying relevant people with explicit links to associated people or implicit links generated when users are creating metadata.

12. A method according to claim 8,

characterized by comprising means for using the viewed and edited metadata from identified users or sessions as a basis for collaborative filtering analysis, and/or means for applying searches to all metadata associated with all documents in a result set, and/or means for information discovery on attributes explicitly represented in the metadata associated with all documents in a result set, and/or means for information discovery on attributes computed from the metadata associated with all documents in a result set.

13. A method according to claim 12,

characterized by said means further comprises location-based search and discovery by associating the metadata object with a location value annotating the corresponding document.

14. A search engine capable of supporting and implementing the method according to any of the preceding claims in information access or search systems, wherein the search engine is applied to accessing, searching, retrieving and analyzing information from content repositories available over data communication networks, including extranets and intranets, and presenting search and analysis results for end users, wherein the search engine comprises at least a core search engine, a content application programming interface connected to the core search engine via content analysis stage, and a query application programming interface connected to the core search engine via respective content analysis and query analysis stages, and wherein said search engine comprises a metadata server connected with or forming a part of the core search engine for storing, editing and presenting metadata to end users.

15. A search engine according to claim 14,

characterized in comprising an alert engine connected in parallel with the core search engine and between a respectively the content analysis stage and the stages for query and result analysis.

3. A method according to claim 3,

characterized by the document subsection corresponding to a scope or tag in a XML representation of the document, or said document subsection corresponding to a specific time or a time interval inside audio or video.