System, method, and user interface for a search engine based on multi-document summarization

• US 8,239,358 B1
• Filed: 01/30/2008
• Issued: 08/07/2012
• Est. Priority Date: 02/06/2007
• Status: Active Grant

• Current Assignee: Dmitri Soubbotin
• Original Assignee: Dmitri Soubbotin
• Inventors: Soubbotin, Dmitri
• Primary Examiner(s): Wu, Yicun
• Assistant Examiner(s): Mamillapalli, Pavan
• Application Number:
  US12/023,014
• Time in Patent Office: 1,651 Days
• Field of Search: 707/707, 707/706
• US Class Current: 707/706
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/345 Summarisation for human users

Abstract

A method for searching multiple documents on a computer system includes steps for sending a query to a system core where the query is passed to a search component for searching the documents. The system core in turn receives results from the search component indicating related documents to the query and passes to a summarization component a specified number of the results. The summarization component processes related documents corresponding to the specified number of results to produce a multi-document summary. The system core receives the summary from the summarization component. The multi-document summary is received from the system core.

Claims

1. A method for searching multiple documents on a computer system, the method comprising steps for:

sending a query to a system core where said query is passed to a search component for searching the documents, said system core in turn receiving results from said search component indicating related documents to said query and passing to a summarization component a specified number of said results where said summarization component processes related documents corresponding to said specified number of results to produce a multi-document summary comprising a digest, on a topic specified by said query, produced by analyzing syntax and semantics of said related documents, extracting sentences and text fragments of said related documents where each of said sentences and text fragments further comprises an attached reference link to at least one related document from which each has been extracted, and combining said sentences and text fragments with their attached reference links into said digest, said system core receiving said multi-document summary from said summarization component;

receiving from said system core said multi-document summary for display as a product of said query;

steps for displaying said multi-document summary;

steps for receiving from said system core said results from said search component and displaying said results in addition to displaying said multi-document summary;

steps for extracting and displaying semantic concepts to a user; and

steps for enabling the user to use the presentation by the displaying steps to refine the query, to navigate to certain parts of said multi-document summary, to view a one of said related documents by choosing a one of said attached reference links in said direst, to view a one of said related documents by choosing a one of said results from said search component, to request another multi-document summary, to exclude one or more sentence fragments from said requested another multi-document summary, to exclude one or more related documents from said requested another multi-document summary, to include one or more related documents excluded in said specified number of said results, to choose a display preference for said attached reference links, to modify or regroup the contents of said multi-document summary, and to save the currently displayed multi-document summary and to rate said multi-document summary feedback to said system core.

2. The method as recited in claim 1, further comprising steps for sending to said system core parameters for limiting a length of said multi-document summary regardless of said specified number of results.

3. The method as recited in claim 1, further comprising steps for rating said multi-document summary by said user for feedback to said summarization component.

4. The method as recited in claim 1, further comprising steps for manipulating said multi-document summary.

5. A method for searching multiple documents on a computer system, the method comprising:

steps for sending a query to a system core; and

steps for receiving from said system core a multi-document summary comprising a digest on a topic specified by said query for display as a product of said query, said digest comprising a combination of sentences and text fragments extracted from related documents returned by said query where each of said sentences, text fragments further comprises an attached reference link to at least one related document from which each has been extracted and

steps for displaying said multi-document summary;

steps for receiving from said system core said results from said search component and displaying said results in addition to displaying said multi-document summary;

steps for extracting and displaying semantic concepts to a user; and

steps for enabling the user to use the presentation by the displaying steps to refine the query, to navigate to certain parts of said multi-document summary, to view a one of said related documents by choosing a one of said attached reference links in said digest, to view a one of said related documents by choosing a one of said results from said search component, to request another multi-document summary, to exclude one or more sentence fragments from said requested another multi-document summary, to exclude one or more related documents from said requested another multi-document summary, to include one or more related documents excluded in said specified number of said results, to choose a display preference for said attached reference links, to modify or regroup the contents of said multi-document summary, and to save the currently displayed multi-document summary and to rate said multi-document summary for feedback to said system core.

6. The method as recited in claim 5, further comprising steps for enabling a user to specify a way attached reference links are displayed in said multi-document summary.

7. A system for searching multiple documents on a computer system, the system comprising:

means for interfacing with a user;

means for searching the documents using a query from the user;

means for generating a multi-document summary comprising a digest having references to related documents from results from said search component, said digest comprising a combination of sentences and text fragments extracted from said related documents where each of said sentences and text fragments further comprises an attached reference link to at least one related document from which each has been extracted;

means for presenting said multi-document summary to the user as a product of said query;

means for receiving from said system core said results from said search component and presenting said results in addition to presenting said multi-document summary;

means for extracting and displaying semantic concepts to the user; and

means for enabling the user to use the presentation by the presenting steps to refine the query, to navigate to certain parts of said multi-document summary, to view a one of said related documents by choosing a one of said attached reference links in said direst, to view a one of said related documents by choosing a one of said results from said search component, to request another multi-document summary, to exclude one or more sentence fragments from said requested another multi-document summary, to exclude one or more related documents from said requested another multi-document summary, to include one or more related documents excluded in said specified number of said results, to choose a display preference for said attached reference links, to modify or regroup the contents of said multi-document summary, to save the currently displayed multi-document summary, and to rate said multi-document summary for feedback to said generating means.

8. The system as recited in claim 7, further comprising means for focusing said multi-document summary on a topic.

9. The system as recited in claim 7, further comprising means for using dictionaries of words or phrases in said generating means for analyzing syntax and semantics of said results.

10. The system as recited in claim 7, further comprising means for selecting a language of said query.

11. The system as recited in claim 7, further comprising means for said user to obtain a source document from said multi-document summary.

12. The system as recited in claim 7, further comprising means for selecting results to pass to said generating means.

13. The system as recited in claim 7, further comprising means for specifying parameters of said multi-documentsummary.

14. The system as recited in claim 7, further comprising means for scoring relevance of semantic concepts when generating said multi-document summary.

15. The system as recited in claim 7, further comprising means for manipulating said multi-document summary.

16. The system as recited in claim 7, further comprising means for said user to rate, share;

comment;

edit;

print or save said multi-document summary.

17. The system as recited in claim 7, further comprising means for integrating said searching means and said generating means.