Apparatus and methods for a computer-aided decision-making system

• US 7,130,836 B2
• Filed: 09/28/2004
• Issued: 10/31/2006
• Est. Priority Date: 02/05/1999
• Status: Expired due to Fees

• Current Assignee: XFI Corp.
• Original Assignee: XFI Corp.
• Inventors: Ghaed, Ali, Grosser, John T.
• Primary Examiner(s): Hirl, Joseph P.
• Application Number:
  US10/954,057
• Publication Number:
  US 20050086187A1
• Time in Patent Office: 763 Days
• Field of Search: 706/47, 706/52, 706/46, 705/26
• US Class Current: 706/47
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

A computer-aided decision-making system and method that is applicable to a variety of decision-making contexts and applications such as, but not limited to, automobile or home purchase decisions. The computer-aided decision-making system provides immediate, useful, and relevant information to a person in a decision-making context, overcoming common human cognitive problems that occur in decision-making, and enabling consumer purchases in an on-line sales environment. In particular, aspects of the invention that aid a person in decision-making include, but are not limited to: managing all the sub-decisions, educating the decision-maker, highlighting the most important sub-decisions, offering the most viable proposals for evaluation, distinguishing significant differences between proposals, supplying various evaluation tools, preventing blind spots, assisting the decision-maker'"'"'s memory, gauging the progress of the decision process, and learning about the decision maker from the decision process.

Claims

1. A computer-aided decision-making system, comprising:

a rules-based analysis engine employing a plurality of rules, said rules being used for selecting and scoring and ranking a plurality of choices;

a user interface operable to accept user-provided information and selections and responses to system inquiries, said user interface indicating system-generated reports and proposals and feedback, said user interface comprising a plurality of advocates selectable by a user for assisting in a single choice and controlled by said rules-based analysis engine to aid the user in making the single choice, said advocates being abstract personalities embodied in software and representing points of view with respect to a decision to be made on the single choice, said points of view including relatively stronger positions on some issues and relatively weaker positions on other issues; and

wherein said rules-based analysis engine accepts said user-provided information and presents through said user interface choices to aid the user in making a decision, said choices being at least one of commented on and chosen by said advocates.

2. A system of claim 1, wherein said user interface further comprises an upper pane and a lower pane for displaying a plurality of reports and decision gauges.

3. The system of claim 1, wherein said plurality of advocates each posses articulate behavior by printed message, sound or movement in response to said user'"'"'s selection of a choice, said articulate behavior indicating the point of view of a corresponding one of said advocates.

4. The system of claim 1, wherein said rules-based analysis engine includes a plurality of databases, said databases comprising:

a merchant/products database containing detailed product or service information; and

a shopper profile database containing individual shopper or purchaser information pertaining to at least one of personal characteristics and purchasing behavior and product/service preferences.

5. The system of claim 4, wherein said rules-based analysis engine comprises:

a residual knowledge database for maintaining errata and modifications to information contained in said merchant/products database; and

a user decision document database capable of storing decision state information responsive to the user requesting to store a current decision state for later use.

6. The system of claim 4, wherein said rules-based analysis engine further comprises:

a plurality of database descriptions for indexing of information contained in said databases;

a plurality of system rules/facts that are applicable to a plurality of decision domains; and

a plurality of application rules/facts that are specific to a particular decision domain.

7. The system of claim 1, wherein said rules-based analysis engine is a fuzzy logic system.

8. A computer-aided decision making system, comprising:

a web browser;

a server-side application/web server, wherein said web browser and said server-side application/web server form a distributed computing client-server system;

a plurality of applets for running a rules-based analysis engine, rules of said rules-based analysis engine being used for selecting scoring and ranking a plurality of choices presented by said client-server system, said scoring and ranking retaining all of said plurality of choices without removal of lower ranking choices; and

an interface having a first portion for displaying system-made choices and a second portion for displaying user-made choices, said first portion being distinct from said second portion, said second portion accepting the user-made choices as user preferences, said user choices being promoted from said first portion to said second portion;

wherein said client-server system aids a user browsing on the web in making a decision.

9. A computer-aided decision making system, comprising:

a web browser;

a server-side application/web server, wherein said web browser and said server-side application/web server form a distributed computing client-server system;

a plurality of applets for running a rules-based analysis engine, rules of said rules-based analysis engine being used for selecting scoring and ranking a plurality of choices presented by said client-server system, said scoring and ranking retaining all of said plurality of choices without removal of lower ranking choices; and

an interface having a first portion for displaying system-made choices and a second portion for displaying user-made choices, said second portion accepting the user-made choices as user preferences, said user choices being promoted from said first portion to said second portion;

wherein said client-server system aids a user browsing on the web in making a decision, wherein said plurality of applets provide a plurality of advocates giving advise to the user to aid in making a single decision, said plurality of advocates providing comments on at least said user-made choices, said advocates being abstract personalities embodied in software and representing points of view with respect to said decision, said points of view including relatively stronger positions on some issues and relatively weaker positions on other issues.

10. A computer implemented method for assisting a person in making a decision, comprising:

decomposing a choice into a plurality of subchoices, wherein said subchoices represent different dimensions of a decision space;

determining a plurality of potential proposals for each said subchoice;

ranking said subchoices according to a range of general to specific subchoices and presenting said ranking to a user for random access at any position along the general to specific ranking;

ranking said plurality of potential proposals in accordance with a user'"'"'s selection among said plurality of subchoices;

ordering said proposals using presentation to indicate a relative ranking of said proposals;

presenting all of said plurality of proposals in ranking order to the user without removing proposals from said plurality of proposals; and

accepting a selection by the user of any proposal of said plurality of proposals regardless of a position of a chosen proposal in said ranking order.

11. The method of claim 10, wherein said step of decomposing a choice into a plurality of subchoices further comprises:

retrieving system-provided subchoices specific to a decision domain from a set of application rules/facts;

accepting user input regarding said subchoices via an input;

aggregating said subchoices into distinct facets based upon type of information;

establishing a multidimensional hierarchical relationship among said facets, wherein said facets are ordered within a dimension according to increasing specificity to an ultimate selected proposal; and

accepting a user selection of an ultimate proposal prior to receiving input from the user for each of said facet so that the decision process is ended prior to consideration of each facet.

12. The method of claim 10, wherein said step of determining a plurality of potential proposals further comprises:

retrieving system-provided values for each said subchoice from a set of databases;

accepting user-provided values via said input for at least one of said subchoices;

calculating an overall weighted score for each subchoice by use of an analysis engine combining said system-provided values and said user-provided values to establish boundaries to said potential proposals based on said system provided values and said user provided values;

generating a plurality of proposals in accordance with said weighted score, aid generated plurality of proposals including proposals both within said boundary and outside said boundary; and

accepting promotion of a proposal by the user from outside said boundary to inside said boundary.

13. The method of claim 12, wherein said step of calculating further comprises:

adjusting value of said system-provided values to user selected values;

predicting by look-ahead an adjustment of a ranking of a choice in order to solicit user feedback for that choice when a determination is made that such user feedback would assist a user in converging to a selected choice.

14. A computer readable media that causes a computer-aided decision-making system to aid a user in making a choice, comprising performing the steps of:

retrieving system-provided subchoices specific to a decision domain from a set of application rules/facts;

accepting user input regarding said subchoices via an input;

aggregating said subchoices into distinct facets based upon type of information;

establishing a multidimensional hierarchical relationship among said facets, wherein said facets are ordered within a dimension according to increasing specificity to an ultimate selected proposal to define ordered facets; and

permitting user access to ones of said facets without regard to a position of a currently accessed facet relative to a previously accessed facet of said ordered facets, said user access permitting user promotion of subchoices.

15. The computer readable media of claim 14, wherein a plurality of advocates are presented to said user to aid in selecting a proposal, said plurality of advocates each being an abstract personality embodied in software and representing a point of view with respect to the decision, said point of view including relatively stronger positions on some issues and relatively weaker positions on other issues.

16. The computer readable media of claim 15, wherein said plurality of advocates articulate their potentially different opinions relating to said point of view by one of printed message and sound and movement.

17. A computer-aided decision making system for deciding among a plurality of choices, comprising:

a rules-based analysis engine;

a user interface accepting input from a user and providing output including results of application of said rules-based analysis engine to inputs of the user;

a plurality of advocates having predetermined mutually different perspectives relating to said choices, said plurality of advocates applying said predetermined mutually different perspectives to said choices and presenting corresponding different perspectives to the user, said plurality of advocates being abstract personalities embodied in software, said abstract personalities having relatively stronger positions on some issues related to said decision and relatively weaker positions on other issues related to said decision.

18. A computer-aided decision making system as claimed in claim 17, further comprising:

an advocate control operable so that the user selectively applies said predetermined perspectives of said plurality of advocates to said choices.

19. A computer-aided decision making system as claimed in claim 17, wherein said plurality of advocates provide said predetermined perspectives to the user in an unsolicited manner.

20. A computer-aided decision making system as claimed in claim 19, wherein said unsolicited perspectives of said plurality of advocates are triggered by an input by the user that represents a position contrary to a position represented by a prior input by the user.

21. A computer-aided decision making system as claimed in claim 20, wherein said plurality of advocates have mutually different strengths of positions on issues relating to said decision, and said plurality of advocates are presented to the user on said user interface ordered according to said strengths of said positions of said advocates.

22. A computer-aided decision making system as claimed in claim 19, further comprising:

an advocate mute operable by the user to selectively mute a presentation of a perspective from said plurality of advocates.

23. A computer-aided decision making system as claimed in claim 19, further comprising:

a threshold for advocate responses which inhibits responses from said plurality of advocates when a difference between an expected choice and a user-entered choice is below said threshold.

24. A computer-aided decision making system as claimed in claim 17, wherein said plurality of advocates present educational information to the user relating so said choices.

25. A computer implemented method of sorting proposals, comprising the steps of:

providing a plurality of choices for a user;

accepting input from the user;

applying a rules-based analysis to said plurality of choices from said input of the user to assign a rank to each of said choices; and

presenting an ordered sort of all choices of said plurality of choices according to said rules-based analysis;

accepting designations of choices preferred by the user to a short list;

fixing sort values in said ordered sort of choices designated to said short list;

accepting further input from the user, said further input including input that is inconsistent with the designations by the user, said inconsistent input being accepted; and

resorting choices based on said further input without resorting choices designated to said short list.

26. A computer implemented method of sorting proposals, comprising the steps of:

providing a plurality of choices to a plurality of users;

receiving input from the plurality of users;

applying a rules-based analysis to said plurality of choices from said input of the user to assign a rank to each of said choices;

presenting an ordered sort of all choices of said plurality of choices according to said rules-based analysis;

providing a plurality of advocates, each of said advocates being an abstract personality embodied in software and representing a point of view with respect to a single decision, said point of view including strong positions on some issues relating to said decision and weak positions on other issues relating to said decision, one of said advocates being a user advocate associated with at least one of the users, said user advocate representing the relatively stronger positions and relatively weaker positions of said at least one user relating to said proposals to others of the users.

27. A method of rules based analysis in a computer aided decision-making system in which rules operate on logical entities of a decision frame, comprising the steps of:

sorting hierarchically arranged facets horizontally in order of increasing specificity to thereby provide;

upper panes containing input or proposal choices reflecting user values, said input or proposal choices each being one of choices and other decision frames; and

lower panes containing input or proposal choices reflecting system generated surrogate values in which choices are sorted vertically in order of decreasing importance to decision outcome, said values being provided by a user/decision maker in said upper pane and by the system in said lower pane;

providing a plurality of advocates each representing a different predefined point of view, said point of view being an encapsulation of a different set of inter-choice rules and a different set of surrogate values and which are capable of informing the user/decision maker of input that meets a criteria of being at least one of inconsistent and incomplete and inadequate; and

providing gauges which inform the decision maker of a degree to which further input is capable of influencing a decision state.

28. A method of sorting proposals, comprising the steps of:

logically grouping a set of input parameter choices into one or more subsets called facets based on relative generality and specificity of the parameters to a decision topic at hand; and

logically sorting said facets from general to specific wherein the facets sorted to a more general relate to a decision topic at hand and being sorted to a left hand and the facets sorted to a more specific relate to the decision and being sorted to a right hand, a most specific facet containing final proposal choices.

29. A method of claim 28, wherein said choices are sorted within a facet in decreasing order of influence to a proposal sort.

30. A method of claim 28, wherein said facets comprise the following:

user information, topics, goals, requirements, and proposals.

31. A method of claim 28, wherein one or more of the more general facets provided at a left hand portion are made available to other applications or sub-choices.

32. A method of claim 28, wherein the system generated sort order of proposal choices are manually altered by a user through promotion into a logical upper pane, causing the system to resort all current lower pane choices in all facets.

33. A method of claim 28, further comprising the step of:

grouping proposal choices into a facet, said choices being sorted according to decreasing desirability based on a current choice state in all other facets.

34. A method of claim 28, wherein choices are sorted within facets from the choices that are most influential in the final facet and proposal sort to the choices that are least influential to the final facet and proposal sort, said most influential being sorted to a top of the facet and said least influential being sorted to a bottom of the facet.

35. A method of claim 28, further comprising the step of:

applying a look-ahead function to elevate a ranking of choices in the lower pane and other facets based on choices promoted to the upper pane in the current facet.

36. A method of claim 28, further comprising the step of:

logically dividing choices and proposal choices within a facet into an initially empty promotion/upper pane set and lower pane set, said lower pane set being initially sorted according to a predefined point of view or surrogate value scheme.

37. A method of claim 28, further comprising the step of:

initially sorting choices and setting said choices to reflect a predefined point of view comprising a set of default or surrogate values, said predefined point of view representing an abstract personality or stereotype.

38. A method of claim 28, further comprising the step of:

selecting a plurality of predefined points of view or value schemes by a user to best approximate the users own point of view or value scheme, said plurality of points of view being blended together to affect a sorting of choices within a facet, said sorting bringing more influential choices to a top of the list within each facet, said more influential choices being more suitable proposal choices.

39. A method of claim 28, further comprising the step of:

accepting at least one point of view from the actual input of other live users in a decision state dataset available to the user, whereby collaborative decision is possible within a topic.

40. A method of claim 28, further comprising the step of:

accepting user input as refined values of choices and proposal choices by promoting said choices to an upper pane subset within a facet to provide a promotion list, said promotion list having enhanced influence on and overall sort of both choices and proposal choices and ranking higher in influence than un-promoted choices within a facet.

41. A method of claim 28, further comprising the step of:

performing a re-sort upon promotion of a proposal choice into an upper pane list to cause resorting of an entire decision state.

42. A method of claim 28, wherein facets are accessed in multiple iterations and random order, each iteration refining the sorting of proposal choices to conform more closely to a user'"'"'s value scheme as represented by the user'"'"'s current choices, advocates, and collaborators.

43. A method of claim 28, wherein act of promotion causes all choices and proposal choices in all facets to be re-sorted in light of changed influence due to the choice promotion.

44. A method of claim 28, wherein an act of promoting a choice recursively invokes another instance of the method of sorting proposals to aid the user in deciding on a choice value within the current decision topic, a proposal choice of the invoked method furnishing the choice value of the invoking method.

45. A method of claim 28, wherein promotion of one or more choices within a facet causes a re-sorting of choices within each logically adjacent facet, each of which causes a resorting of its adjacent facets, including ultimately a logically rightmost proposal facet, resulting in every choice promotion causing all other choices to be re-sorted in an order of maximum influence and relevance, and all proposal choices to be re-sorted in an order most closely conforming to the user'"'"'s actual inputs and the influence of selected advocates.

46. A method of claim 28, wherein resorting is by transposition of rows in a facet (n) into columns of a facet (n+1).