Method and system for ascribing a reputation to an entity as a rater of other entities

• US 6,895,385 B1
• Filed: 11/10/2000
• Issued: 05/17/2005
• Est. Priority Date: 06/02/2000
• Status: Expired due to Fees

• Current Assignee: Open Ratings, Inc. (Dun & Bradstreet Corp.)
• Original Assignee: Open Ratings
• Inventors: Tkach, Dmitry, Zacharia, Giorgos C.
• Primary Examiner(s): Hafiz, Tariq
• Assistant Examiner(s): Stimpak, Johnna
• Application Number:
  US09/710,008
• Time in Patent Office: 1,649 Days
• Field of Search: 705/7, 705/8, 705/10, 705/11
• US Class Current: 705/7.29
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data

Abstract

A method and system for ascribing a reputation of an entity as a rater of other entities is provided. A first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, are provided. The second ratings are combined to produce a first combined rating. The first rating is compared to the first combined rating to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating. A resulting rater reputation is generated based at least in part on the first rating predictability.

Claims

1. A method of determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, the method comprising computer-implemented acts of:

(A) combining the second ratings to produce a first combined rating, comprising an act of (i) calculating a rating average of the second ratings, the rating average being the first combined rating;

(B) comparing the first rating to the first combined rating to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating; and

(C) generating a resulting rater reputation based at least in part on the first rating predictability.

2. The method of claim 1, wherein the first rating is indicative of a rating of a rated entity by the first entity at a first time, and each second rating is indicative of a rating of the rated entity provided by another entity at a respective time after the first time.

3. The method of claim 2, further comprising an act of:

(D) receiving a value indicating a number of most recent ratings, wherein the most recent ratings include the second ratings,

wherein act (C) comprises

(1) determining a number of the second ratings;

(2) dividing the number of second ratings plus one by the number of most recent ratings to produce a ratio; and

(3) weighting the predictability with the ratio.

4. The method of claim 1, further comprising:

(D) determining a ratee reputation of the rated entity based at least in part on the first rating.

5. The method of claim 1, wherein each entity that provided a second rating has a respective second rater reputation as a rater of entities, wherein act (A)(i) comprises, for each second rating:

applying a first weighting to the second rating, wherein the first weighting is a positive function of the second rater reputation of the entity that provided the second rating relative to the second rater reputations of the other second entities, such that the calculated average is a weighted average.

6. The method of claim 5, wherein the first rating is indicative of a rating of a rated entity by the first entity at a first time, and each second rating is indicative of a rating of the rated entity provided by another entity at a time after the first time, and

wherein each second rating has a position in an order, the position corresponding to a temporal closeness of the time that the second rating was provided to the first time relative to the temporal closeness of the times that the other second ratings were provided to the first time, wherein act (A)(i) further comprises:

for each second rating, applying a second weighting to the second rating, wherein the second weighting is a function of the position in the order of the second rating, such that the temporally closer the time the second rating was provided to the first time relative to the temporal closeness of the times of the other second rating to the first time then the greater the weight the second rating has in calculating the average and determining the rater reputation.

7. The method of claim 1, wherein the first rating is indicative of a rating of a rated entity by the first entity at a first time, and each second rating is indicative of a rating of the rated entity provided by another entity at a time later than the first time, and

wherein each second rating has a position in an order, the position corresponding to a temporal closeness of the time that the;

second rating was provided to the first time relative to the temporal closeness of the times that the other second ratings were provided to the first time, wherein act (A)(i) further comprises;

for each second rating, applying a second weighting to the second rating, wherein the second weighting is a function of the position in the order of the second rating, such that the temporally closer the time the second rating was provided to the first time relative to the temporal closeness of the times of the other second ratings to the first time then the greater the weight the second rating has in calculating the average and the determining the rater reputation.

8. The method of claim 1, wherein also provided is an initial rater reputation indicative of an initial rater reputation of the first entity, the method further comprising acts of:

(D) subtracting the first rating predictability from an initial rater reputation of the first entity to produce a first rater reputation modification; and

(E) adding to the initial rater reputation a first rater reputation adjustment that is a positive function of the first rater'"'"'s reputation modification to produce the resulting rater reputation.

9. The method of claim 8, wherein each entity that provided a second rating has a second rater reputation as a rater of entities, the method further comprising acts of:

(F) for one or more of the second ratings, combining the second rater reputations of the entities that provided the one or more of the second ratings to produce a combined rater reputation; and

(G) scaling the first rater reputation modification by the combined rater reputation such that a magnitude of the first rater reputation adjustment corresponds to the combined rater reputation.

10. The method of claim 9, wherein act (F) comprises:

(i) calculating a rater reputation average of the second rater reputations of the entities that provided the one or more second ratings, wherein the rater reputation average is the combined rater reputation.

11. The method of claim 8, wherein each first and second rating has a value greater than zero, and wherein the first rating is a first-time rating provided by the first entity for any entity, the method further comprising the act of:

(F) before performing act (D), initializing the initial rater reputation of the first entity to zero such that the first entity has a lowest rater reputation prior to providing the first rating.

12. The method of claim 8, wherein the first rating is one of a plurality of ratings of one or more other entities provided by the first entity that have not previously been used to determine the rater reputation of the first entity, the method further comprising acts of:

(G) repeating acts (A)-(E) for each of the remaining ratings of the plurality of ratings to produce a first final resulting rater reputation;

(H) setting the initial rater reputation equal to the first resulting rater reputation; and

(I) repeating act (A)-(E) for each of the plurality of ratings.

13. The method of claim 8, further comprising an act of:

(F) reducing the first rater reputation modification by a change rate factor that controls a rate at which the rater reputation of the first entity can be changed.

14. The method of claim 8, wherein further provided is a third rating having a value indicative of a rating of a second entity by the first entity, and one or more fourth ratings, each fourth rating having a value indicative of a rating of the second entity provided by another entity, the method further comprising acts of:

(F) combining the values of the fourth ratings to produce a second combined rating having a value;

(G) comparing the value of the third rating to the value of the second combined rating to produce a second rating predictability of the third rating, the second rating predictability increasing as a magnitude of a difference between the value of the third rating and the value of the second combined rating decreases and decreasing as the magnitude of the difference increases;

(H) subtracting the second rating predictability from the resulting rater reputation to produce a second rater reputation modification; and

(I) adding to the resulting rater reputation a second rater reputation adjustment that is a function of the second rater reputation modification to produce a new resulting rater reputation.

15. The method of claim 1, wherein act (B) comprises acts of:

(i) determining a difference between the first rating and the first combined rating;

(ii) dividing the determined difference by a ratee rating deviation of the rated entity, the ratee rating deviation defining a deviation of ratings by other entities of the rated entity from a mean rating; and

(iii) determining the first rating predictability from the divided difference.

16. The method of claim 15, wherein act (B)(iii) comprises:

applying a Gaussian distribution function to the divided difference to produce the first rating predictability.

17. The method of claim 1, further comprising acts of:

(D) receiving a third rating of a second entity provided by the first entity; and

(E) determining a ratee reputation of the second entity based at least in part on the resulting rater reputation of the first entity.

18. The method of claim 17, further comprising an act of:

(F) determining whether to transact with the second entity based on the determined ratee reputation of the second entity.

19. The method of claim 17, further comprising an act of:

(F) determining a price to pay for a good or service of the second entity based at least in part on the determined ratee reputation of the second entity.

20. The method of claim 17, further comprising an act of:

(F) determining a price to pay for insuring a quality of a good or service of the second entity based at least in part on the determined ratee reputation of the second entity.

21. The method of claim 1, further comprising acts of:

(D) determining a rater reputation deviation of the first entity, the rater reputation deviation defining a deviation of two or more rating predictabilities of ratings provided by the first entity of other entities from an expected rating predictability of ratings provided by the first entity; and

(E) assessing a reliability of the resulting rater reputation based at least in part on the rater rating predictability deviation.

22. The method of claim 21, wherein each entity that provided a second rating has a second rater reputation as a rater of entities, and wherein each rating predictability is weighted as a positive function of an average rater reputation corresponding to the rating predictability, each average rater reputation defined as an average of the second reputations of one or more second ratings corresponding to the rating predictability.

23. The method of claim 1, wherein no two or more second ratings are made by a same entity.

24. The method of claim 1, wherein no two or more second ratings are made by a same entity within a particular temporal period.

25. The method of claim 1, wherein the first entity, the other entities, and the rated entity are all members of a population, and each first and second rating of the rated entity is a part of a set of multiple ratings of the rated entity by the first and second entity, respectively, and wherein the population is divided into a plurality of clusters in correlation to the statistical similarities of the multiple ratings, the method further comprising an act of:

(D) selecting the other entities that provided the second ratings based on the selected other entities and the first entity being members of a same cluster.

26. A system for determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, the system comprising:

a rater reputation generator to combine the second ratings to produce a first expected rating;

to compare the first rating to the first combined rating to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating, to generate a resulting rater reputation based at least in part on the first rating predictability, and to provide as output the rater reputation of the first entity, wherein the rater reputation generator comprises an expected rating generator to generate the first expected rating by calculating a rating average of the second ratings.

29. The system of claim 26, wherein the first rating is indicative of a rating of a rated entity by the first entity at a first time, and each second rating is indicative of a rating of the rated entity provided by another entity at a respective time after the first time.

30. The system of claim 26, further comprising:

a ratee reputation generator to determine a ratee reputation of the rated entity based at least in part on the first rating.

31. The system of claim 26, wherein each entity that provided a second rating has a respective second rater reputation as a rater of entities, and wherein the expected rating generator is further operative to apply, for each second rating, a first weighting to the second rating, wherein the first weighting is based on the second rater reputation of the entity that provided the second rating relative to the second rater reputations of the other second entities, such that the calculated average is a weighted average.

32. The system of claim 26, wherein also provided is an initial rater reputation indicative of an initial rater reputation of the first entity, the system further comprising:

a reputation modification generator to subtract the first rating predictability from an initial rater reputation of the first entity to produce a first rater reputation modification; and

an adder to add, to the initial rater reputation, a first rater reputation adjustment that is a function of the first rater reputation modification to produce the resulting rater reputation.

33. The system of claim 32, wherein each first and second rating has a value greater than zero, and the first rating is a first-time rating provided by the first entity for any entity, and wherein the system is operative to initialize the initial rater reputation of the first entity to zero, before the reputation modification generator subtracts the first-rating predictability from the initial rater reputation, such that the first entity has a lowest rater reputation prior to providing the first rating.

34. The system of claim 32, further comprising:

a reputation adjustment generator to reduce the first rater reputation modification by a change rate factor that controls a rate at which the rater reputation of the first entity can be changed.

35. The system claim 26, wherein no two or more second ratings are made by a same entity.

36. The system of claim 26, wherein no two or more second ratings are made by a same entity within a particular temporal period.

37. The system of claim 26, wherein the first entity, the other entities, and the rated entity are all members of a population, and each first and second rating of the rated entity is a part of a set of multiple ratings of the rated entity by the first and second entity, respectively, and wherein the population is divided into a plurality of clusters in correlation to the statistical similarities of the multiple ratings,

wherein the rater reputation generator is operative to select the other entities that provided the second ratings based on the selected other entities and the first entity being members of a same cluster.

27. A system for determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, the system comprising:

means for combining the second ratings to produce a first expected rating, wherein the means for combining comprises means for generating the first expected rating by calculating a rating average of the second ratings;

means for comparing the first rating to the first combined rating numerical to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating; and

means for generating a resulting rater reputation based at least in part on the first rating predictability.

38. The system of claim 27, wherein the first rating is indicative of a rating of a rated entity by the first entity at a first time, and each second rating is indicative of aerating of the rated entity provided by another entity at a respective time after the first time.

39. The system of claim 27, further comprising:

means for determining a ratee reputation of the rated entity based at least in part on the first rating.

40. The system of claim 27, wherein each entity that provided a second rating has a respective second rater reputation as a rater of entities, wherein the means for generating the first expected rating comprises:

means for applying, for each second rating, a first weighting to the second rating, wherein the first weighting is based on the second rater reputation of the entity that provided the second rating relative to the second rater reputations of the other second entities, such that the calculated average is a weighted average.

41. The system of claim 27, wherein also provided is an initial rater reputation indicative of an initial rater reputation of the first entity, the system further comprising:

means for subtracting the first predictability from an initial rater reputation of the first entity to produce a first rater reputation modification; and

means for adding to the initial rater reputation first rater reputation adjustment that is a function of the first rater reputation modification to produce the resulting rater reputation.

42. The system of claim 41, wherein each first and second rating has a value greater than zero, and wherein the first rating is a first-time rating provided by the first entity for any entity, the system further comprising:

means for initializing, before subtracting the first predictability from the initial rater reputation, the initial rater reputation of the first entity to zero such that the first entity has a lowest rater reputation prior to providing the first rating.

43. The system of claim 41, further comprising:

means for reducing the first rater reputation modification by a change rate factor that controls a rate at which the rater reputation of the first entity can be changed.

44. The system of claim 27, wherein no two or more second ratings are made by a same entity.

45. The system of claim 27, wherein no two or more second ratings are made by a same entity within a particular temporal period.

46. The system of claim 27, wherein the first entity, the other entities, and the rated entity are all members of a population, and each first and second rating of the rated entity is a part of a set of multiple ratings of the rated entity by the first and second entity, respectively, and wherein the population is divided into a plurality of clusters in correlation to the statistical similarities of the multiple ratings, the system further comprising:

means selecting the other entities that provided the second ratings based on the selected other entities and the first entity being members of a same cluster.

28. A computer program product, comprising:

a computer readable medium; and

computer readable signals stored on the computer readable medium that define instructions that, as a result of being executed by a computer, instruct the computer to perform a method of determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, the method comprising acts of;

(A) combining the second ratings to produce a first expected rating wherein the act (A) comprises an act of (i) generating the first expected rating by calculating a rating average of the second ratings;

(B) comparing the first rating to the first combined rating to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating; and

(C) generating a resulting rater reputation based at least in part on the first rating predictability.

47. The computer program product of claim 28, wherein the first rating is indicative of a rating of a rated entity by the first entity at a first time, and each second rating is indicative of a rating of the rated entity provided by another entity at a respective time after the first time.

48. The computer program product of claim 28, wherein the method further comprises:

(D) determining a ratee reputation of the rated entity based at least in part on the first rating.

49. The computer program product of claim 28, wherein each entity that provided a second rating has a respective second rater reputation as a rater of entities, wherein act (A)(i) comprises, for each second rating:

applying a first weighting to the second rating, wherein the first weighting is based on the second rater reputation of the entity that provided the second rating relative to the second rater reputations of the other second entities, such that the calculated average is a weighted average.

50. The computer program product of claim 28, wherein also provided is an initial rater reputation indicative of an initial rater reputation of the first entity, the method further comprising acts of:

(D) subtracting the first predictability from an initial rater reputation of the first entity to produce a first rater reputation modification; and

(E) adding to the initial rater reputation a first rater reputation adjustment that is a function of the first rater reputation modification to produce the resulting rater reputation.

51. The computer program product of claim 50, wherein each first and second rating has a value greater than zero, and wherein the first rating is a first-time rating provided by the first entity for any entity, the method further comprising an act of:

(F) before performing act (D), initializing the initial rater reputation of the first entity to zero such that the first entity has a lowest rater reputation prior to providing the first rating.

52. The computer program product of claim 50, further comprising an act of:

(F) reducing the first rater reputation modification by a change rate factor that controls a rate at which the rater reputation of the first entity can be changed.

53. The computer program product of claim 28, wherein no two or more second ratings are made by a same entity.

54. The computer program product of claim 28, wherein no two or more second ratings are made by a same entity within a particular temporal period.

55. The computer program product of claim 28, wherein the first entity, the other entities, and the rated entity are all members of a population, and each first and second rating of the rated entity is a part of a set of multiple ratings of the rated entity by the first and second entity, respectively, and wherein the population is divided into a plurality of clusters in correlation to the statistical similarities of the multiple ratings, the method further comprising an act of:

(D) selecting the other entities that provided the second ratings based on the selected other entities and the first entity being members of a same cluster.

56. A method of determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another respective entity, the method comprising computer-implemented acts of:

(A) generating an expected rating by combining the second ratings, wherein the act (A) comprises an act of (i) generating the expected rating by calculating a rating average of the second ratings;

(B) generating a predictability of the first rating by comparing the first rating to the expected rating; and

(C) generating a resulting rater reputation based at least in part on the first predictability.

57. The method of claim 56, wherein the predictability represents a closeness of the first rating to the expected rating.

58. The method of claim 56, wherein the predictability is a negative function of a magnitude of a difference between the first rating and the expected rating.

59. A system for determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another respective entity, the system comprising:

a rater reputation generator to generate an expected rating by combining the second ratings, to generate a predictability of the first rating by comparing the first rating to the expected rating, and to generate a resulting rater reputation based at least in part on the first predictability, wherein the rater reputation generator comprises an expected rating generator to generate the expected rating by calculating a rating average of the second ratings.

60. The system of claim 59, wherein the predictability represents a closeness of the first rating to the expected rating.

61. The system of claim 59, wherein the predictability is a negative function of a magnitude of a difference between the first rating and the expected rating.

62. A system for determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another respective entity, the system comprising:

means for generating an expected rating by combining the second ratings, wherein the means for generating comprises means for generating the expected rating by calculating a rating average of the second ratings;

means for generating a predictability of the first rating by comparing the first rating to the expected rating; and

means for generating a resulting rater reputation based at least in part on the first predictability.

63. The system of claim 62, wherein the predictability represents a closeness of the first rating to the expected rating.

64. The system of claim 62, wherein the predictability is a negative function of a magnitude of a difference between the first rating and the expected rating.

65. A computer-readable medium having computer-readable signals stored thereon that define instructions that, as a result of being executed by a computer, instruct the computer to perform a method of determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another respective entity, the method comprising acts of:

(A) generating an expected rating by combining the second ratings, wherein the act (A) comprises an act of (i) generating the expected rating by calculating a rating average of the second ratings;

(B) generating a predictability of the first rating by comparing the first rating to the expected rating; and

(C) generating a resulting rater reputation based at least in part on the first predictability.

66. The computer-readable medium of claim 65, wherein the predictability represents a closeness of the first rating to the expected rating.

67. The computer-readable medium of claim 65, wherein the predictability is a negative function of a magnitude of a difference between the first rating and the expected rating.

68. A method of determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, and wherein the first rating is indicative of a rating of a rated entity by the first entity at a first time, and each second rating is indicative of a rating of the rated entity provided by another entity at a respective time after the first time, the method comprising computer-implemented acts of:

(A) combining the second ratings to produce a first combined rating;

(B) comparing the first rating to the first combined rating to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating;

(C) receiving a value indicating a number of most recent ratings, wherein the most recent ratings include the second ratings; and

(D) generating a resulting rater reputation based at least in part on the first rating predictability, comprising;

(1) determining a number of the second ratings;

(2), dividing the number of second ratings plus one by the number of most recent ratings to produce a ratio; and

(3) weighting the predictability with the ratio.

69. A method of determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, and wherein also provided is an initial rater reputation indicative of an initial rater reputation of the first entity, the method comprising computer-implemented acts of:

(A) combining the second ratings to produce a first combined rating;

(B) comparing the first rating to the first combined rating to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating;

(C) generating a resulting rater reputation based at least in part on the first rating predictability, including;

(i) subtracting the first rating predictability from an initial rater reputation of the first entity to produce a first rater reputation modification; and

(ii) adding to the initial rater reputation a first rater reputation adjustment that is a positive function of the first rater'"'"'s reputation modification to produce the resulting rater reputation.

70. A method of determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, the method comprising computer-implemented acts of:

(A) combining the second ratings to produce a first combined rating;

(B) comparing the first-rating to the first combined rating to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating, wherein act (B) comprises acts of;

(i) determining a difference between the first rating and the first combined rating;

(ii) dividing the determined difference by a ratee rating deviation of the rated entity, the ratee rating deviation defining a deviation of ratings by other entities of the rated entity from a mean rating; and

(iii) determining the first rating predictability from the divided difference, wherein act (B)(iii) comprises applying a Gaussian distribution function to the divided difference to produce the first rating predictability; and

(C) generating a resulting rater reputation based at least in part on the first rating predictability.

71. A method of determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, the method comprising computer-implemented acts of:

(A) combining the second ratings to produce a first combined rating;

(B) comparing the first rating to the first combined rating to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating;

(C) generating a resulting rater reputation based at least in part on the first rating predictability;

(D) determining a rater reputation deviation of the first entity, the rater reputation deviation defining a deviation of two or more rating predictabilities of ratings provided by the first entity of other entities from an expected rating predictability of ratings provided by the first entity; and

(E) assessing a reliability of the resulting rater reputation based at least in part on the rater rating predictability deviation.

72. The method of claim 71, wherein each entity that provided a second rating has a second rater reputation as a rater of entities, and wherein each rating predictability is weighted as a positive function of an average rater reputation corresponding to the rating predictability, each average rater reputation defined as an average of the second reputations of one or more second ratings corresponding to the rating predictability.

73. A system for determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, and wherein also provided is an initial rater reputation indicative of an initial rater reputation of the first entity, the system comprising:

a rater reputation generator to combine the second ratings to produce a first combined rating;

to compare the first rating to the first combined rating to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating, to generate a resulting rater reputation based at least in part on the first rating predictability, and to provide as output the rater reputation of the first entity, wherein the rater reputation generator includes;

subtracting the first rating predictability from an initial rater reputation of the first entity to produce a first rater reputation modification; and

adding to the initial rater reputation a first rater reputation adjustment that is a positive function of the first rater'"'"'s reputation modification to produce the resulting rater reputation.

74. The system of claim 73, wherein each first and second rating has a value greater than zero, and the first rating is a first-time rating provided by the first entity for any entity, and wherein the system is operative to initialize the initial rater reputation of the first entity to zero, before the reputation modification generator subtracts the first rating predictability from the initial rater reputation, such that the first entity has a lowest rater reputation prior to providing the first rating.

75. The system of claim 73, wherein the rater reputation generator comprises:

a reputation adjustment generator to reduce the first rater reputation modification by a change rate factor that controls at a rate at which the rater reputation of the first entity can be changed.

76. A system for determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, wherein also provided is an initial rater reputation indicative of an initial rater reputation of the first entity, the system comprising:

means for combining the second ratings to produce a first combined rating;

means for comparing the first rating to the first combined rating numerical to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating;

means for generating a resulting rater reputation based at least in part on the first rating predictability;

means for subtracting the first predictability from an initial rater reputation of the first entity to produce a first rater reputation modification; and

means for adding to the initial rater reputation a first rater reputation adjustment that is a function of the first rater reputation modification to produce the resulting rater reputation.

77. The system of claim 76, wherein each first and second rating has a value greater than zero, and wherein the first rating is a first-time rating provided by the first entity for any entity, the system further comprising:

means for initializing, before subtracting the first predictability from the initial rater reputation, the initial rater reputation of the first entity to zero such that the first entity has a lowest rater reputation prior to providing the first rating.

78. The system of claim 76, further comprising:

means for reducing the first rater reputation modification by a change rate factor that controls a rate at which the rater reputation of the first entity can be changed.

79. A computer program product, comprising:

a computer readable medium; and

computer readable signals stored on the computer readable medium that define instructions that, as a result of being executed by a computer, instruct the computer to perform a method of determining a rater reputation of a first entity as a rater of other entities, wherein provided is a first rating indicative of a rating of a rated entity by the first entity, and one or more second ratings, each second rating indicative of a rating of the rated entity provided by another entity, and wherein also provided is an initial rater reputation indicative of an initial rater reputation of the first entity, the method comprising acts of;

(A), combining the second ratings to produce a first combined rating;

(B) comparing the first rating to the first combined rating to produce a first rating predictability of the first rating, the first rating predictability being a negative function of a magnitude of a difference between the first rating and the first combined rating;

(C) generating a resulting rater reputation based at least in part on the first rating predictability, including;

(i) subtracting the first rating predictability from an initial rater reputation of the first entity to produce a first rater reputation modification; and

(ii) adding to the initial rater reputation a first rater reputation adjustment that is a positive function of the first rater'"'"'s reputation modification to produce the resulting rater reputation.

80. The computer program product of claim 79, wherein each first and second rating has a value greater than zero, and wherein the first rating is a first-time rating provided by the first entity for any entity, the method further comprising an act of:

(F) before performing act (D), initializing the initial rater reputation of the first entity to zero such that the first entity has a lowest rater reputation prior to providing the first rating.

81. The computer program product of claim 79, further comprising an act of:

(F) reducing the first rater reputation modification by a change rate factor that controls a rate at which the rater reputation of the first entity can be changed.