System and method for dynamic data-mining and on-line communication of customized information

• US 6,266,668 B1
• Filed: 08/04/1999
• Issued: 07/24/2001
• Est. Priority Date: 08/04/1998
• Status: Expired due to Term

• Current Assignee: Mantra Technologies Incorporated
• Original Assignee: Dryken Technologies, Inc.
• Inventors: Vanderveldt, Ingrid V., Black, Christopher Lee
• Primary Examiner(s): Powell, Mark R.
• Assistant Examiner(s): Khatri, Anil
• Application Number:
  US09/366,590
• Time in Patent Office: 720 Days
• Field of Search: 707/10, 707/100, 706/12, 706/14, 706/15, 706/16
• 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

Abstract

A method and system for dynamically searching databases in response to a query is provided by the present invention. More specifically, a system and method for dynamic data-mining and on-line communication of customized information. This method includes the steps of first creating a search-specific profile. This search-specific profile is then input into a data-mining search engine. The data-mining search engine will mine the search-specific profile to determine topic of interests. These topics of interest are output to at least one search tool. These search tools match the topics of interest to at least one destination data site wherein the destination data sites are evaluated to determine if relevant information is present in the destination data site. Relevant information is filtered and presented to the user making the inquiry.

Claims

1. A method for assigning relevance of inter-linked objects within a database using an artificial neural network (“

ANN”

) to provide a customized search tool for a user, comprising;

searching said database for one or more inter-linked objects satisfying at least one or more relevance metrics;

assigning, with said ANN, a weight to each of said inter-linked objects; and

identifying to said user each of said one or more weighted inter-linked objects having an assigned weight greater than some predetermined threshold value.

2. The method of claim 1, further comprising the step of training said ANN to assign said weight to each of said inter-linked objects based on said user'"'"'s historical search preferences.

3. The method of claim 1, wherein said database is a partially or a fully indexed database.

4. The method of claim 1, wherein said inter-linked objects are data sites, or links to data sites.

5. The method of claim 1, wherein said relevance metrics are satisfied with reference to a user text query.

6. The method of claim 5, wherein said one or more relevance metrics comprise a first relevance metric, a second relevance metric and a third relevance metric.

7. The method of claim 5, wherein said first metric is based on frequency of common word occurrences between said text query and said inter-linked objects, wherein said second metric is based upon associations between a set of user profile information and said inter-linked objects, and wherein said third relevance metric is based upon whether inter-linked objects having assigned relevance metrics are inter-linked with each other.

8. The method of claim 1, wherein said weight comprises a sum of individual weights assigned by said ANN for each of said relevance metrics satisfied by each of said inter-linked objects.

9. The method of claim 1, wherein said identifying step further comprises organizing by weight said weighted interlinked objects having an assigned weight greater than some predetermined threshold value.

10. A method for dynamically searching a database, in response to a textual query, to provide a customized search tool for a user, comprising the steps of:

(a) comparing said textual query against said database'"'"'s existing static database index to identify, using an Artificial Neural Network (“

ANN”

), a first set of one or more relevant inter-linked objects using one or more relevance metrics;

(b) comparing said textual query against one or more secondary inter-linked objects linked to by one or more of said inter-linked objects in said first set to identify, using said ANN, a second set of relevant inter-linked objects using said one or more relevance metrics; and

(c) repeating step (b) for each subsequent identified set of relevant inter-linked objects a preset number of times.

11. The method of claim 10, wherein steps (a) and (b) further comprise the steps of:

searching said existing database index or said one or more secondary inter-linked objects, respectively, for one or more inter-linked objects satisfying at least one of said one or more relevance metrics;

assigning, by said ANN, a weight to each of said inter-linked objects to represent its relevance; and

populating said first set, or said second set with each of said one or more weighted inter-linked objects having an assigned weight greater than some predetermined threshold value.

12. The method of claim 11, further comprising the step of training said ANN to assign said weight to each of said inter-linked objects based on said user'"'"'s historical search preferences.

13. The method of claim 11, further comprising the step of identifying to a user each of said one or more weighted inter-linked objects having an assigned weight greater than some predetermined threshold value.

14. The method of claim 10, wherein said database is a partially or a fully indexed database.

15. The method of claim 10, wherein said first set and said second set of inter-linked objects comprise sets of data sites, or sets of links to data sites.

16. The method of claim 10, wherein said one or more relevance metrics comprise a first relevance metric, a second relevance metric and a third relevance metric.

17. The method of claim 16, wherein said first metric is based on frequency of common word occurrences between said text query and said inter-linked objects, wherein said second metric is based upon associations between a set of user profile information and said inter-linked objects, and wherein said third relevance metric is based upon whether inter-linked objects having assigned relevance metrics are inter-linked with each other.

18. The method of claim 17, wherein said user profile information comprises demographic information provided or inferred from the said user'"'"'s responses to a registration questionnaire, augmented by explicit or implicit user feedback.

19. The method of claim 10, wherein said weight comprises a sum of individual weights assigned by said ANN for each of said relevance metrics satisfied by each of said inter-linked objects.

20. The method of claim 10, wherein said identifying step further comprises organizing by weight said weighted inter-linked objects having an assigned weight greater than some predetermined threshold value.

21. The method of claim 10, wherein said textual query is user supplied.

22. The method of claim 10, wherein said relevant inter-linked objects are links to predictive models for modeling the subject matter of said textual query.

23. The method of claim 22, wherein said predictive models comprise models for prediction of stock market future prices, medical condition diagnosis, wine selection to accompany a dinner entree, cross sale, add-on sale and other e-commerce product recommendation applications.

24. The method of claim 10, wherein said database comprises links to partially indexed HTML-formatted documents available on the internet and wherein said index is an existing ‘

static’

web search index.

25. A system for assigning relevance of inter-linked objects within a database using an artificial neural network (“

ANN”

) to provide a customized search tool for a user, comprising;

instructions for searching said database for one or more inter-linked objects satisfying at least one of one or more relevance metrics;

instructions for assigning, with said ANN, a weight to each of said inter-linked objects; and

instructions for identifying to said user each of said one or more weighted inter-linked objects having an assigned weight greater than some predetermined threshold value.

26. The system of claim 25, further comprising instructions for training said ANN to assign said weight to each of said inter-linked objects based on said user'"'"'s historical search preferences.

27. The system of claim 25, wherein said database is a partially or a fully indexed database.

28. The system of claim 25, wherein said inter-linked objects are data sites, or links to data sites.

29. The system of claim 25, wherein said relevance metrics are satisfied with reference to a user text query.

30. The system of claim 29, wherein said one or more relevance metrics comprise a first relevance metric, a second relevance metric and a third relevance metric.

31. The system of claim 29, wherein said first metric is based on frequency of common word occurrences between said text query and said inter-linked objects, wherein said second metric is based upon associations between a set of user profile information and said inter-linked objects, and wherein said third relevance metric is based upon whether inter-linked objects having assigned relevance metrics are inter-linked with each other.

32. The system of claim 25, wherein said weight comprises a sum of individual weights assigned by said ANN for each of said relevance metrics satisfied by each of said inter-linked objects.

33. The system of claim 25, wherein said instructions for identifying further comprise instructions for organizing by weight said weighted inter-linked objects having an assigned weight greater than some predetermined threshold value.