Self-learning and self-personalizing knowledge search engine that delivers holistic results

• US 6,397,212 B1
• Filed: 03/04/1999
• Issued: 05/28/2002
• Est. Priority Date: 03/04/1999
• Status: Expired due to Term

• Current Assignee: Hanger Solutions, LLC (IP Investments Group LLC), On-Offline
• Original Assignee: Peter Biffar
• Inventors: Biffar, Peter
• Primary Examiner(s): Alam, Hosain T.
• Application Number:
  US09/262,953
• Time in Patent Office: 1,181 Days
• Field of Search: 707/5,6,10 705/7-11,35-38,14,26
• US Class Current: 707/704
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: Y10S 707/914 Video
• CPC Class Codes: Y10S 707/917 Text
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

A search engine provides intelligent multi-dimensional searches, in which the search engine always presents a complete, holistic result, and in which the search engine presents knowledge (i.e. linked facts) and not just information (i.e. facts). The search engine is adaptive, such that the search results improve over time as the system learns about the user and develops a user profile. Thus, the search engine is self personalizing, i.e. it collects and analyzes the user history, and/or it has the user react to solutions and learns from such user reactions. The search engine generates profiles, e.g. it learns from all searches of all users and combines the user profiles and patterns of similar users. The search engine accepts direct user feedback to improve the next search iteration One feature of the invention is locking/unlocking, where a user may select specific attributes that are to remain locked while the search engine matches these locked attributes to all unlocked attributes. The user may also specify details about characteristics, provide and/or receive qualitative ratings of an overall result, and introduce additional criteria to the search strategy or select a search algorithm. Additionally, the system can be set up such that it does not require a keyboard and/or mouse interface, e.g. it can operate with a television remote control or other such human interface.

Claims

1. A search apparatus, comprising:

an item database which contains an item for which a search is to be performed, wherein each item has an associated characteristic, wherein similar items are structured along a same characteristic, and wherein it is not necessary that all items of an item group have all the same characteristics;

a logic module for determining those items that must be searched for in said item database;

wherein said logic module searches said item database;

a presentation module for displaying a search result; and

means for allowing a user to lock certain characteristics for said search by said logic module;

wherein said logic module attempts to fulfill higher level, more important characteristics first, violating lower level characteristics if there is differentiation in importance of characteristics.

2. The apparatus of claim 1, further comprising:

a user database for storing user searches, wherein said user database stores historically most popular characteristic or characteristics which may include those which fit to a profile of a user, those which are known through available user data, those which are previously provided directly or indirectly by said user, or those which are deducted from earlier selection of characteristics.

3. The apparatus of claim 1, said presentation module further acting as an input for additional searches.

4. The apparatus of claim 1, wherein each item stored in said item database comprises a complete product or service.

5. The apparatus of claim 1, wherein information describing a characteristic of an item is attached to each characteristic, wherein such information can be any length and any format.

6. The apparatus of claim 1, wherein said characteristics are differentiated by importance in hierarchies.

7. The apparatus of claim 1, wherein an item is a sub-item that is defined as a combination of other items, and wherein said sub-items are further characteristics of said item.

8. The apparatus of claim 1, wherein each item is dynamically generated and/or created based on logic module rules contained in said logic module, and wherein a dynamically generated item is created by combining separate characteristics to create one complete item that comprises desired characteristics.

9. The apparatus of claim 1, wherein each item is a combination of some characteristics which are fixed and other characteristics that can be added dynamically.

10. The apparatus of claim 1, said logic module further comprising logic module rules that describe how to arrive at a characteristic.

11. The apparatus of claim 1, wherein an item, characteristic, and information describing a characteristic searched for can reside on different databases and/or Web sites.

12. The apparatus of claim 1, wherein characteristics can reside on multiple databases and/or Web sites.

13. The apparatus of claim 1, wherein an item can be tagged on-the-fly.

14. The apparatus of claim 1, wherein said presentation module further comprises any of:

a description of an item;

a characteristic;

information;

an action; and

a locking function that allows a user to lock in characteristics which are desired.

15. The apparatus of claim 14, wherein said description can be text based, a picture, video, sound, or a combination thereof.

16. The apparatus of claim 14, wherein said characteristics and said information describe an item shown along with said characteristics.

17. The apparatus of claim 14, wherein said action allows search parameters for a next search to be changed by a user.

18. The apparatus of claim 17, wherein said action may be executed by any of a pull down menu, yes and no questions, multiple choice questions, and ratings.

19. The apparatus of claim 1, further comprising:

a new search button that allows a user to select a next iteration to find a better fitting solution.

20. The apparatus of claim 1, further comprising:

a rating function for allowing a user to rate a search.

21. The apparatus of claim 1, further comprising:

means for allowing a user to move away from a next best fitting iteration to a totally new starting point in the search.

22. The apparatus of claim 1, further comprising:

means for allowing a user to select or build/customize an algorithm.

23. The apparatus of claim 1, wherein said presentation module is configured such that a display is not needed.

24. The apparatus of claim 1, wherein said logic module is adaptive, wherein search results improve over time as said logic module learns about a user and develops a user profile.

25. The apparatus of claim 24, wherein user profile information is acquired directly from either of questions asked of said user or a third party.

26. The apparatus of claim 1, wherein said logic module is self personalizing, wherein said logic module collects and analyzes history about a user, has said user react to solutions, and learns from those reactions.

27. The apparatus of claim 1, wherein said logic module is also profiling, wherein said logic module learns from all searches of all users and combines a user profile and patterns of similar users.

28. The apparatus of claim 1, wherein said item database is a virtual database in which items reside on many remote Web sites and/or databases;

and wherein a smart agent finds desired items in said virtual database.

29. The apparatus of claim 1, wherein said logic module substitutes a next best solution for locked characteristics.

30. The apparatus of claim 1, wherein said logic module bundles search results with other items.

31. The apparatus of claim 1, wherein said logic module provides a save search result function that allows a user to begin an alternative search.

32. The apparatus of claim 1, wherein said presentation module displays several search results in a tabular form.

33. The apparatus of claim 1, wherein said logic module allows a user to include and/or exclude any of various characteristics.

34. The apparatus of claim 1, wherein said logic module allows various characteristics to be detailed by said user.

35. The apparatus of claim 1, wherein said logic module allows said user to determine the logical relationship to each other of details of characteristics and/or characteristics themselves.

36. A search method, comprising the steps of:

providing an item database which contains an item for which a search is to be performed, wherein each item has an associated characteristic, wherein similar items are structured along a same characteristic, and wherein it is not necessary that all items of an item group have all the same characteristics;

providing a logic module for determining those items that must be searched for in said item database;

wherein said logic module searches said item database;

providing a presentation module for displaying a search result; and

providing means for allowing a user to lock certain characteristics for said search by said logic module;

wherein said logic module attempts to fulfill higher level, more important characteristics first, violating lower level characteristics if there is differentiation in importance of characteristics.

37. The method of claim 36, further comprising the step of:

providing a user database for storing user searches, wherein said user database stores historically most popular characteristics or such characteristics as those which fit to a profile of a user, those which are known through available user data, those which are previously provided directly or indirectly by said user, or those which are deducted from earlier selection of characteristics.

38. The method of claim 36, wherein said presentation module acts as an input for additional searches.

39. The method of claim 36, further comprising the steps of:

initiating a search;

using a search engine to select an initial search result as a starting point;

listing the most important characteristic for a subsequent search result;

wherein a next response is a search result that fits well to a new set of characteristic; and

wherein said search engine uses an internal algorithm to fit in missing information if said user has not specified said characteristics using preselected, default, or most appropriate values; and

said user optionally selecting a next result/search algorithm.

40. The method of claim 39, wherein said starting point is context based.

41. The method of claim 39, wherein said initial search result is selected at random, preset default characteristics or results are provided, or said initial search result is selected from historically most popular characteristics or such characteristics as those which fit to a profile of a user, those which are known through available user data, those which are previously provided directly or indirectly by said user, or those which are deducted from earlier selection of characteristics.

42. The method of claim 39, further comprising any of the steps of:

selecting a result; and

pointing out a violation to said user if certain characteristics are locked, or if user chosen characteristics conflict with each other;

wherein said user can reject a result offered and a new result is then generated violating a different characteristic without changing said locking.

43. The method of claim 39, further comprising the step of:

generating results using a conjoint analysis algorithm, wherein at least some of the solutions generated are not a best fitting solution, but a solution selected to help assess a user'"'"'s preferences and relative importance of said characteristics.

44. The method of claim 39, further comprising the step of:

ordering said results to a fit of characteristics and allows the user to scroll through them in sequential order.

45. The method of claim 39, further comprising the step of:

displaying a comparison table of a present and a previous search.

46. The method of claim 39, further comprising the step of:

asking said user to provide an indication of importance of characteristics when presented with a solution.

47. The method of claim 39, further comprising the step of:

providing a user database that contains all historic data about users and searches.

48. The method of claim 47, wherein said data includes information about a user where such information is collected based on searches conducted, third party information acquired about said user, information about profiles of users or user groups against which a profile of a specific user can be measured, and information about what items a user having a similar profile has obtained.

49. The method of claim 47, wherein said database comprises a plurality of databases that are distributed over a network and connected to each other.

50. The method of claim 36 wherein said item database is a virtual database in which items reside on many remote Web sites and/or databases;

and

wherein a smart agent finds desired items in said virtual database.

51. The method of claim 37, wherein said user database is a virtual database in which information about user reside on many remote Web sites and/or databases;

and

wherein a smart agent finds desired user information in said virtual database.

52. The method of claim 36, wherein said logic module substitutes a next best solution for locked characteristics.

53. The method of claim 36, wherein said logic module bundles primary search results with secondary items.

54. The method of claim 36, wherein said logic module provides a save search result function that allows a user to begin an alternative search.

55. The method of claim 36, wherein said logic module allows a user to include and/or exclude any of various characteristics.

56. The method of claim 36, wherein said logic module allows various characteristics to be detailed by said user.

57. The method of claim 36, wherein said logic module allows said user to determine the logical relationship to each other of details of characteristics and/or characteristics themselves.