초록 ▼

A system for internet navigation using soft hyperlinks is disclosed, in connection with an illustrative information retrieval system with which it may be used. The navigation tool provides freedom to move through a collection of electronic documents independent of any hyperlink which has been insert

A system for internet navigation using soft hyperlinks is disclosed, in connection with an illustrative information retrieval system with which it may be used. The navigation tool provides freedom to move through a collection of electronic documents independent of any hyperlink which has been inserted within an HTML page. A user can click on any term in a document page, not only those that are hyperlinked. For example, when a user clicks on an initial word within the document, the disclosed system employs a search engine in the background to retrieve a list of related terms. In an illustrative embodiment, a compass-like display appears with pointers indicating the first four terms returned by the search engine. These returned terms have the highest degree of correlation with the initial search term in a lexical knowledge base that the search engine constructs automatically. The disclosed system allows the user to move from the current document to one of a number of document lists which cover different associations between the initial word clicked on by the user and other terms extracted from within the retrieved list of related terms. The disclosed system may further allow the user to move to a document that is considered most related to the initial word clicked on by the user, or to a list of documents that are relevant to a phrase or paragraph selection indicated by the user within the current page.

대표청구항 ▼

1. A computer-implemented method for navigating through a set of electronic documents, comprising:capturing an indication of an initial term displayed in connection with a document, wherein the initial term is one of a plurality of terms displayed in connection with the document, wherein the plurali

1. A computer-implemented method for navigating through a set of electronic documents, comprising:capturing an indication of an initial term displayed in connection with a document, wherein the initial term is one of a plurality of terms displayed in connection with the document, wherein the plurality of terms includes hyperlinked and non-hyperlinked terms; issuing an initial search request of the set of electronic documents, via a search engine, using an initial search query consisting of the initial term, the initial search request causing the search engine to perform a correlation analysis of the set of documents compared with the initial search query; receiving a plurality of related terms from the search engine as an initial search result, the related terms determined from the correlation analysis of the set of documents; selecting a predetermined number of the received plurality of related terms and presenting the selected related terms graphically; capturing an indication of at least one of the plurality of related terms; and issuing, responsive to the indication of the at least one of the related terms, at least one secondary search request, wherein a secondary search query associated with the secondary search request includes the indicated related term and the initial term; storing a plurality of secondary search result document weights; and retrieving an information file associated with a first one of the document weights and displaying a portion of the retrieved information file containing a first occurrence of the initial term. 2. The method of claim 1, further comprising retrieving one or more of a plurality of information files associated with respective ones of the plurality of secondary search result document weights, and displaying portions of the plurality of information files containing first occurrences of the initial term.3. The method of claim 1 wherein the capturing the indication of the initial term displayed in connection with the document further comprises capturing a non-hyperlinked term, and wherein the initial term includes the non-hyperlinked term.4. The method of claim 1 wherein the issuing the secondary search request includes forming a logical AND expression between the indicated related term and the initial term, and wherein the secondary search query includes the logical AND expression between the indicated related term and the initial term.5. The method of claim 1 wherein the capturing the indication of the initial term displayed in connection with the document further comprises capturing a phrase, and wherein the initial term includes the indicated phrase.6. The method of claim 1 wherein the capturing the indication of the initial term displayed in connection with the document further comprises capturing a paragraph, and wherein the initial term includes the indicated paragraph.7. The method of claim 1 wherein the capturing the indication of the initial term displayed in connection with the document further comprises capturing one or more indicated documents, and wherein the initial term includes the indicated documents.8. The method of claim 1 wherein the selecting the predetermined number of the received plurality of related terms includes associating a relevance level with each one of the related terms selected from the received plurality of related terms.9. The method of claim 8 wherein each relevance level reflects a lexical correlation between the initial term used in the initial search request and the respective selected one of the received plurality of related terms.10. The method of claim 1 wherein the electronic documents comprise Web pages accessible through the World Wide Web.11. The method of claim 1 wherein the electronic documents comprise information files accessible through the Internet.12. The method of claim 1 wherein the correlation analysis of the set of documents performed by the search engine uses Latent Semantic Regression.13. The method of claim 1, the receiving the plurality of related terms from the search engine as an initial search result further comprising receiving the plurality of related terms from the search engine as an initial search result, wherein at least two of the received related terms occur in at least two documents that do not contain co-occurrences of the two related terms.14. The method of claim 1, the receiving the plurality of related terms from the search engine as an initial search result further comprising receiving the plurality of related terms from the search engine as an initial search result, the related terms being sorted based upon correlation to the initial term.15. The method of claim 1, the storing the plurality of secondary search result document weights further comprising storing the plurality of secondary search result document weights, the document weights being sorted in decreasing order.16. The method of claim 1, the presenting the selected related terms graphically further comprising presenting the selected related terms as part of a compass-like navigation tool.17. A navigation system for examining a set of electronic documents, comprising:a search engine structured to perform a correlation analysis of a query and the set of electronic documents and determine related terms, each related term an indicator of a set of documents that is related to the query; a graphic navigation tool structured to receive and display a plurality of secondary terms; capture an indication of at least one of the plurality of displayed secondary terms; issue a secondary search request that includes the indicated secondary term and the initial term in a secondary search query; receive a plurality of indicators of documents that are related to the secondary search query, the indicators of the related documents having associated document weights; retrieve an information file associated with a first one of the document weights; and display a portion of the retrieved information file containing a first occurrence of the initial term; and a user interface structured to capture an indication of an initial term from a plurality of terms displayed in connection with the document, wherein the displayed plurality of terms includes hyperlinked and non-hyperlinked terms; issue an initial search request of the set of electronic documents to the search engine, using an initial search query consisting of the initial term; receive a plurality of related terms from the search engine as an initial search result, the related terms determined from the correlation analysis; select a predetermined number of the received plurality of related terms; and invoke the graphic navigation tool with the selected related terms as the plurality of secondary terms to display. 18. The navigation system of claim 17 wherein the graphic navigation tool retrieves one or more of a plurality of information files associated with respective ones of the plurality of document weights, and displays portions of the plurality of information files containing first occurrences of the initial term.19. The navigation system of claim 17 wherein the captured initial term includes the non-hyperlinked term.20. The navigation system of claim 17 wherein the secondary search query is formed using a logical AND expression between the indicated related term and the initial term.21. The navigation system of claim 17 wherein the captured indication of the initial term includes a phrase.22. The navigation system of claim 17 wherein the captured indication of the initial term includes a paragraph.23. The navigation system of claim 17 wherein the captured indication of the initial term includes a document.24. The navigation system of claim 17 wherein a relevance level is associated with each one of the related terms selected from the received plurality of related terms.25. The navigation system of claim 24 wherein each relevance level reflects a lexical correlation between the initial term used in the initial search request and the respective selected one of the received plurality of related terms.26. The navigation system of claim 17 wherein the electronic documents comprise information files accessible through the Internet.27. The navigation system of claim 17 wherein the correlation analysis of the set of documents performed by the search engine uses Latent Semantic Regression.28. The navigation system of claim 17 wherein at least two of the related terms determined by the search engine indicate at least two documents that do not contain co-occurrences of the two related terms such that the at least two documents are latently related through a third related term.29. A computer-implemented method for navigating through a corpus of documents stored in a network of computer systems, comprising:displaying a document with a plurality of terms and receiving an indication of an initial one of the plurality of terms; automatically causing a search engine to perform a search using an initial search query that contains the indicated initial term, the search performing a correlation analysis of the corpus of documents relative to the initial term and determining a set of related terms from documents that are directly related to the initial term and from documents that have latent associations with the initial term, each related term associated with a set of documents that are related to the initial term; receiving from the search engine the determined set of related terms and presenting a determined number of the received related terms through a navigation tool; receiving an indication of one of the presented related terms; automatically causing a secondary search to be performed using a secondary search query that includes the initial term and the indicated related term, the secondary search determining a set of weighted indicators to documents; and receiving the determined set of weighted indicators to documents and displaying a portion of one or more documents that are indicated by one or more of the weighted indicators. 30. The method of claim 29 wherein the navigation tool is a compass-like display of the determined number of the received related terms.31. The method of claim 29 wherein each of the received related terms is associated with a relevance level and the presenting the determined number of the receiving related terms includes presenting the associated relevance levels.32. The method of claim 29 wherein the determining the set of related terms from documents that are directly related to the initial term and from documents that have latent associations with the initial term, each related term associated with a set of documents that are related to the initial term further comprises:determining the set of related terms from documents that contain co-occurrences of the initial term and one of the related terms and from documents that contain co-occurrences of at least two of the related terms but not the initial term, each related term thereby associated with a set of documents that are directly and latently related to the initial term. 33. The method of claim 29 wherein each displayed portion of one or more documents that are indicated by one or more of the weighted indicators indicates a first occurrence of the initial term.34. The method of claim 29 wherein the search performs the correlation analysis of the corpus of documents relative to the initial term using a Latent Semantic Regression analysis.35. The method of claim 29 wherein the plurality of terms displayed in the document includes hyper-linked terms and non-hyperlinked terms.36. The method of claim 35 wherein the initial term includes a non-hyperlinked term.37. The method of claim 29 wherein each term comprises one or more words, phrases, documents, figures, and objects.38. The method of claim 29 wherein the navigation tool displays the determined number of the received related terms in an area separate from the document such as a menu, a list, or other graphical presentation.39. The method of claim 29 wherein the network is the Internet.40. The method of claim 29 wherein the corpus of documents includes web pages.41. A computer-readable memory medium containing instructions for controlling a computer processor to navigate through a corpus of documents stored in a network of computer systems, by:displaying a document with a plurality of terms and receiving an indication of an initial one of the plurality of terms; automatically causing a search engine to perform a search using an initial search query that contains the indicated initial term, the search performing a correlation analysis of the corpus of documents relative to the initial term and determining a set of related terms from documents that are directly related to the initial term and from documents that have latent associations with the initial term, each related term associated with a set of documents that are related to the initial term; receiving from the search engine the determined set of related terms and presenting a determined number of the received related terms through a navigation tool; receiving an indication of one of the presented related terms; automatically causing a secondary search to be performed using a secondary search query that includes the initial term and the indicated related term, the secondary search determining a set of weighted indicators to documents; and receiving the determined set of weighted indicators to documents and displaying a portion of one or more documents that are indicated by one or more of the weighted indicators. 42. The memory medium of claim 41 wherein the navigation tool is a compass-like display of the determined number of the received related terms.43. The memory medium of claim 41 wherein the determining the set of related terms from documents that are directly related to the initial term and from documents that have latent associations with the initial term, each related term associated with a set of documents that are related to the initial term further comprises:determining the set of related terms from documents that contain co-occurrences of the initial term and one of the related terms and from documents that contain co-occurrences of at least two of the related terms but not the initial term, each related term thereby associated with a set of documents that are directly and latently related to the initial term. 44. The memory medium of claim 41 wherein each displayed portion of one or more documents that are indicated by one or more of the weighted indicators indicates a first occurrence of the initial term.45. The memory medium of claim 41 wherein the search performs the correlation analysis of the corpus of documents relative to the initial term using a Latent Semantic Regression analysis.46. The memory medium of claim 41 wherein the plurality of terms displayed in the document includes hyper-linked terms and non-hyperlinked terms.47. The memory medium of claim 46 wherein the initial term includes a non-hyperlinked term.48. The memory medium of claim 41 wherein each term comprises one or more words, phrases, documents, figures, and objects.49. The memory medium of claim 41 wherein the navigation tool displays the determined number of the received related terms in an area separate from the document such as a menu, a list, or other graphical presentation.50. A navigation system for determining relevant information from a corpus of documents stored in a network of computer systems, comprising:a search engine that is structured to perform a correlation analysis of the corpus of documents relative to a search query containing an initial term and to determine a set of related terms from documents that are directly related to the initial term and from documents that have latent associations with the initial term, each related term associated with a set of documents that are related to the initial term; and a user interface that is structured to display a document with a plurality of terms and receive an indication of one of the plurality of terms; automatically invoke the search engine using a search query that contains the indicated term and receive from the search engine the determined set of related terms; present a determined number of the received related terms through a navigation tool and receive an indication of one of the presented related terms; automatically cause a secondary search to be performed using a secondary search query that includes the initial term and the indicated related term, the secondary search determining a set of weighted indicators to documents; and receive the determined set of weighted indicators to documents and display a portion of one or more documents that are indicated by one or more of the weighted indicators. 51. The system of claim 50 wherein the navigation tool is a compass-like display of the determined number of the received related terms.52. The system of claim 50 wherein each of the received related terms is associated with a relevance level and the user interface is structured to present the determined number of the received related terms with the associated relevance levels.53. The system of claim 50 wherein the search engine is structured to determine the set of related terms from documents that are directly related to the initial term and from documents that have latent associations with the initial term by determining the set of related terms from documents that contain co-occurrences of the initial term and one of the related terms and from documents that contain co-occurrences of at least two of the related terms but not the initial term.54. The system of claim 50 wherein the user interface is further structured to display each displayed portion of one or more documents that are indicated by one or more of the weighted indicators in a manner that indicates a first occurrence of the initial term.55. The system of claim 50 wherein the search engine performs the correlation analysis of the corpus of documents relative to the initial term using Latent Semantic Regression.56. The system of claim 50 wherein the plurality of terms displayed in the document includes hyper-linked terms and non-hyperlinked terms.57. The system of claim 56 wherein the initial term includes a non-hyperlinked term.58. The system of claim 50 wherein each term comprises one or more words, phrases, documents, figures, and objects.59. The system of claim 50 wherein the navigation tool displays the determined number of the received related terms in an area separate from the document such as a menu, a list, or other graphical presentation.60. The system of claim 50 wherein the network is the Internet.61. The system of claim 50 wherein the corpus of documents includes web pages.