Facility for highlighting documents accessed through search or browsing

• US 20060059138A1
• Filed: 10/21/2005
• Published: 03/16/2006
• Est. Priority Date: 05/25/2000
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Milic-Frayling, Natasa, Sommerer, Ralph
• Granted Patent: US 7,660,813 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

An information highlighting facility assists the user in evaluating relevance of accessed documents to the user'"'"'s information need. The accessed documents may, for example, be identified by a search engine in response to a user query. When accessing documents identified as relevant by a search engine from other networked computers, the facility provides information highlighting to assist the user in determining whether the document is relevant. A model of the user'"'"'s interest, which may include an augmented set of search terms, is used to take into account the general interest of the user as captured by an interest profile and context of use of the computer by the user, or a combination thereof. The model of the user'"'"'s interest is applied to the document text as the document is accessed from its source. The highlighting of information about the document content may include highlighting of the terminology in the text, scrolling of the document to the relevant passages, identification of entity names and entity relations, creation of a document summary and a document thumbnail, etc. In addition, the model can be applied to a set of documents accessed by the user, e.g., to re-rank the top scoring documents from the result set provided to the user by a search engine or some other information providing services.

Claims

1. A computer implemented method of enhancing query results provided independent of a search engine, the method comprising:

sending a query to an independent search engine;

receiving query results from the search engine; and

generating information regarding relevancy of the query results based at least in part upon a user model and independent of the search engine.

2. The method of claim 1, highlighting relevant portions of text in the retrieved documents based upon the information regarding relevancy.

3. The method of claim 1, extracting names from the document and identifying associated links to such names based at least in part upon a user model.

4. The method of claim 3, the names comprise names of people or companies.

5. The method of claim 3, the links are internal to the document.

6. The method of claim 3, the links are external to the document.

7. The method of claim 3, the names are provided in a list next to the query results to help identify the relevance of documents.

8. The method of claim 1, creating a thumbnail view of the document with portions of the view highlighted based on relevancy of corresponding portions of the document, such thumbnail view based at least in part upon a user model.

9. The method of claim 8, the highlighted portions of the thumbnail view correspond to links back to corresponding portions of text in the document.

10. A computer implemented method of enhancing query results provided independent of a search engine, the method comprising:

sending a query to an independent search engine;

creating a context based at least in part upon a user model;

receiving query results from the search engine; and

generating information regarding relevancy of the query results independent of the search engine and based upon the context.

11. The method of claim 10, generating a summary of the document comprising the most relevant sections identified, such summary based at least in part upon the context.

12. The method of claim 11, dividing the document into sections, and generating a relevancy score for each section based upon the context.

13. The method of claim 12, the most relevant section of the document is the section with the highest score.

14. The method of claim 12, one or more sections overlap other sections.

15. The method of claim 12, each section is one of a paragraph, a sentence, and a predetermined number of lines of text.

16. A computer system for enhancing query results provided independent of a search engine, the system comprising:

a module that sends a query to a search engine separate from the computer;

a module that receives query results from the search engine;

a module that retrieves documents identified by the query results;

a module that enhances the query based at least in part upon a user model;

a module that applies the enhanced query to the retrieved documents; and

a module that generates information regarding relevancy of the retrieved documents based at least in part upon the enhanced query and the user model.

17. The system of claim 16, the query is enhanced based on at least one of linguistic analysis, a general interest profile, a model of user interest generated independent of search results.

18. The method of claim 17, the linguistic analysis comprises at least one of syntactic and semantic analysis.

19. The system of claim 16, further containing a module that re-ranks the query results based on information contained in the retrieved documents and upon a user model.

20. The system of claim 16, further containing a module that scrolls to a most relevant portion of a retrieved document based at least in part upon a user model.