US5884282A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Automated collaborative filtering system
Assignee: FELLOWSHIP FILTERING TECHNOLOGIES LLC
•
Filing: 1998-04-09
•
Granted: 1999-03-16
•
Expires: 2018-04-09
Abstract
Abstract An automated collaborative filtering (ACF) system for recommending at least one item to a first user based on similarity in preference of the user as compared with other users. The ACF system stores rating data for items provided by users of the system. Upon request of the first user, the system determines similarity values for the first user as compared with other users that have provided rating data for items that the first user has also rated. Based on the similarity values, a subgroup of users is selected that is then used to provide recommendations to the first user.
Inventors
- 👤 Gary B. Robinson
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-25 22:13
CPC Classifications
G; G06; G06Q; G06Q30/00; G06Q30/02; G06Q30/0201; G06Q30/0204; G06F; G06F16/00; G06F16/30; G06F16/33; G06F16/335; G06Q10/00; G06Q10/02; G06Q10/025; G06Q30/06; G06Q30/0601; G06Q30/0631; Y; Y10; Y10S; Y10S707/00; Y10S707/99941; Y10S707/99943
Claims Analysis
Total Claims: 60
Claim 1
Independent Claim
199 words
A method for selecting an item from a plurality of items to recommend to a first user of a plurality of users, the method comprising: obtaining ratings data for at least some of the items in the plurality of items from first ones of the plurality of users, wherein the ratings data for each of the respective items includes information about the overall ratings distribution for the item; locating a first common item that the first user and a second user in the plurality of users have provided ratings data for; transforming the ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users to provide first transformed ratings data; computing a range of randomized transformed ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users; determining a test statistic for the first common item from the first transformed ratings data and the range of randomized transformed ratings data; and determining a similarity value from the test statistic for the first user with respect to the second user.
Claim 2
Independent Claim
199 words
A method for selecting an item from a plurality of items to recommend to a first user of a plurality of users, the method comprising: obtaining ratings data for at least some of the items in the plurality of items from first ones of the plurality of users, wherein the ratings data for each of the respective items includes information about the overall ratings distribution for the item; locating a first common item that the first user and a second user in the plurality of users have provided ratings data for; transforming the ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users to provide first transformed ratings data; computing a range of randomized transformed ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users; determining a test statistic for the first common item from the first transformed ratings data and the range of randomized transformed ratings data; and determining a similarity value from the test statistic for the first user with respect to the second user.
Claim 3
Dependent Claim
42 words
The method according to claim 1, further including recommending a first item to the first user based on the similarity value of the first user with respect to the second user and the ratings data of items rated by the second user.
Claim 4
Dependent Claim
59 words
The method according to claim 1, wherein for the first common item the test statistic corresponds to a probability that a third user will have a range of randomized transformed ratings that is at least as close to the first transformed ratings data corresponding to the first user as the randomized transformed ratings data corresponding to the second user.
Claim 5
Dependent Claim
13 words
The method according to claim 3, wherein the third user is randomly selected.
Claim 6
Dependent Claim
28 words
The method according to claim 1, further including determining at least one similarity value for the first user with respect to second ones of the plurality of users.
Claim 7
Dependent Claim
41 words
The method according to claim 5, further including determining a subgroup of users for recommending the first item based on the at least one similarity value of the first user with respect to the second ones of the plurality of users.
Claim 8
Dependent Claim
19 words
The method according to claim 1, wherein determining the similarity value includes computing the similarity value using a z-score.
Claim 9
Dependent Claim
20 words
The method according to claim 1, wherein determining the similarity value includes computing the similarity value using a Chi-square distribution.
Claim 10
Dependent Claim
38 words
The method according to claim 1, wherein determining the similarity value includes determining the similarity value from a plurality of test statistics, each corresponding to further items that the first and second users have provided ratings data for.
Claim 11
Dependent Claim
48 words
The method according to claim 1, wherein computing the range of randomized ratings data for the first common item includes computing the range of randomized ratings data by adding a uniformly distributed random number to the ratings data provided by the first ones of the plurality of users.
Claim 12
Dependent Claim
37 words
The method according to claim 1, further including determining further common items that the first and second users have provided ratings for, and assigning a statistical weight to the first common item and the further common items.
Claim 13
Dependent Claim
22 words
The method according to claim 11, wherein assigning the statistical weight further includes assigning the statistical weight using a genetic algorithm technique.
Claim 14
Dependent Claim
20 words
The method according to claim 2, further including providing a review of the first recommended item to the first user.
Claim 15
Dependent Claim
43 words
The method according to claim 2, further including providing a plurality of reviews of the first recommended item by further users to the first user in an order corresponding to respective similarity values of the first user with respect to the further users.
Claim 16
Dependent Claim
24 words
The method according to claim 2, further including allowing the first user to rate reviews of the first recommended item provided by further users.
Claim 17
Dependent Claim
35 words
The method according to claim 15, further including determining a goodness value for at least some of the reviews of the recommended item from the ratings given to the respective review by the first user.
Claim 18
Dependent Claim
24 words
The method according to claim 16, further including ordering the reviews provided to the first user using the goodness value associated with each review.
Claim 19
Dependent Claim
63 words
The method according to claim 1, further including evaluating a recommendation of an item wherein evaluating the recommendation includes recommending a second item that the first and second users have provided ratings for while ignoring the ratings data of the second recommended item provided by the first user, and examining the ratings data of the second recommended item provided by the first user.
Claim 20
Independent Claim
206 words
A system for selecting an item in a plurality of items to recommend to a first user of a plurality of users, comprising: means for obtaining ratings data for at least some of the items in the plurality of items from first ones of the plurality of users, wherein the ratings data for each of the respective items includes information about the overall ratings distribution for the item; means for locating a first common item that the first user and a second user in the plurality of users have provided ratings data for; means for transforming the ratings data for the first common item from the ratings data for the first item provided by the first ones of the plurality of users to provide first transformed ratings data; means for computing a range of randomized transformed ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users; means for determining a test statistic for the first common item from the first transformed ratings data and the randomized transformed ratings data; and means for determining a similarity value from the test statistic for the first user with respect to the second user.
Claim 21
Independent Claim
206 words
A system for selecting an item in a plurality of items to recommend to a first user of a plurality of users, comprising: means for obtaining ratings data for at least some of the items in the plurality of items from first ones of the plurality of users, wherein the ratings data for each of the respective items includes information about the overall ratings distribution for the item; means for locating a first common item that the first user and a second user in the plurality of users have provided ratings data for; means for transforming the ratings data for the first common item from the ratings data for the first item provided by the first ones of the plurality of users to provide first transformed ratings data; means for computing a range of randomized transformed ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users; means for determining a test statistic for the first common item from the first transformed ratings data and the randomized transformed ratings data; and means for determining a similarity value from the test statistic for the first user with respect to the second user.
Claim 22
Dependent Claim
44 words
The system according to claim 19, further including means for recommending a first item to the first user based on the similarity value of the first user with respect to the second user and the ratings data of items rated by the second user.
Claim 23
Dependent Claim
71 words
The system according to claim 19, wherein the means for determining a test statistic for the first common item includes means for determining a test statistic that corresponds to a probability that a third user will have a range of randomized transformed ratings that is at least as close to the first transformed ratings data corresponding to the first user as the randomized transformed ratings data corresponding to the second user.
Claim 24
Dependent Claim
13 words
The system according to claim 21, wherein the third user is randomly selected.
Claim 25
Dependent Claim
25 words
The system according to claim 19, wherein the means for determining a similarity value includes means for determining a similarity value using a statistical z-score.
Claim 26
Dependent Claim
25 words
The system according to claim 19, wherein the means for determining a similarity value includes means for determining a similarity value using a Chi-square distribution.
Claim 27
Independent Claim
218 words
A system for selecting an item in a plurality of items to recommend to a first user of a plurality of users, comprising: an interface for obtaining ratings data for at least some of the items in the plurality of items from first ones of the plurality of users, wherein the ratings data for each of the respective items includes information about the overall ratings distribution for the item; a locator for determining a first common item that the first user and a second user in the plurality of users have provided ratings data for; a transformer for transforming the ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users to provide first transformed ratings data; a ratings processor for computing a range of randomized transformed ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users; a test statistic processor for determining a test statistic for the first common item from the first transformed ratings data and the randomized transformed ratings data; and a similarity value processor for determining a similarity value from the test statistic for the first user with respect to the second user.
Claim 28
Independent Claim
218 words
A system for selecting an item in a plurality of items to recommend to a first user of a plurality of users, comprising: an interface for obtaining ratings data for at least some of the items in the plurality of items from first ones of the plurality of users, wherein the ratings data for each of the respective items includes information about the overall ratings distribution for the item; a locator for determining a first common item that the first user and a second user in the plurality of users have provided ratings data for; a transformer for transforming the ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users to provide first transformed ratings data; a ratings processor for computing a range of randomized transformed ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users; a test statistic processor for determining a test statistic for the first common item from the first transformed ratings data and the randomized transformed ratings data; and a similarity value processor for determining a similarity value from the test statistic for the first user with respect to the second user.
Claim 29
Dependent Claim
46 words
The system according to claim 25, further including a recommendation processor for recommending a first item to the first user based on the similarity value of the first user with respect to the second user and the ratings data of items rated by the second user.
Claim 30
Dependent Claim
61 words
The system according to claim 25, wherein for the first common item the test statistic corresponds to a probability that a randomly chosen third user will have a range of randomized transformed ratings that is at least as close to the first transformed ratings data corresponding to the first user as the randomized transformed ratings data corresponding to the second user.
Claim 31
Dependent Claim
16 words
The system according to claim 25, wherein the similarity value is determined using a statistical z-score.
Claim 32
Dependent Claim
16 words
The system according to claim 25, wherein the similarity value is determined using a Chi-square distribution.
Claim 33
Dependent Claim
34 words
The system according to claim 25, wherein the similarity value is determined from a plurality of test statistics each corresponding to further items that the first and second users have provided ratings data for.
Claim 34
Dependent Claim
35 words
The system according to claim 25, wherein the randomized ratings data for the first common item is generated by adding a uniformly distributed random number to the ratings data provided by the plurality of users.
Claim 35
Dependent Claim
36 words
The system according to claim 25, wherein the locator determines further common items that the first and second user have provided ratings for, and the first common item and the further common items are statistically weighted.
Claim 36
Dependent Claim
24 words
The system according to claim 26, further including a review processor for providing a review of the first recommended item to the first user.
Claim 37
Dependent Claim
45 words
The system according to claim 26, wherein the review processor provides a plurality of reviews of the first recommended item by further users to the first user in an order corresponding to respective similarity values of the first user with respect to the further users.
Claim 38
Dependent Claim
25 words
The system according to claim 25, wherein the review processor allows the first user to rate reviews of the first item provided by further users.
Claim 39
Dependent Claim
32 words
The system according to claim 26, wherein the review processor determines a goodness value for each review of the recommended item from the rating given to each review by the first user.
Claim 40
Dependent Claim
26 words
The system according to claim 36, wherein the review processor orders the reviews provided to the first user using the goodness value associated with each review.
Claim 41
Dependent Claim
62 words
The system according to claim 25, further including an evacuator for evaluating a recommendation of an item by recommending a second item that the first and second users have provided ratings for while ignoring the ratings data of the second recommended item provided by the first user, and examining the ratings data of the second recommended item provided by the first user.
Claim 42
Dependent Claim
208 words
A computer readable medium having computer readable code that causes a computer system to select an item in a plurality of items to recommend to a first user in a plurality of users by: obtaining ratings data for at least some of the items in the plurality of items from first ones of the plurality of users, wherein the ratings data for each of the respective items includes information about the overall ratings distribution for the item; locating a first common item that the first user and a second user in the plurality of users have provided ratings data for; transforming the ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users to provide first transformed ratings data; computing a range of randomized transformed ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users; determining a test statistic for the first common item from the first transformed ratings data and the range of randomized transformed ratings data; and determining a similarity value from the test statistic for the first user with respect to the second user.
Claim 43
Dependent Claim
208 words
A computer readable medium having computer readable code that causes a computer system to select an item in a plurality of items to recommend to a first user in a plurality of users by: obtaining ratings data for at least some of the items in the plurality of items from first ones of the plurality of users, wherein the ratings data for each of the respective items includes information about the overall ratings distribution for the item; locating a first common item that the first user and a second user in the plurality of users have provided ratings data for; transforming the ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users to provide first transformed ratings data; computing a range of randomized transformed ratings data for the first common item from the ratings data for the first common item provided by the first ones of the plurality of users; determining a test statistic for the first common item from the first transformed ratings data and the range of randomized transformed ratings data; and determining a similarity value from the test statistic for the first user with respect to the second user.
Claim 44
Dependent Claim
47 words
The computer-readable medium according to claim 39, further including computer readable code for recommending a first item to the first user based on the similarity value of the first user with respect to the second user and the ratings data of items rated by the second user.
Claim 45
Dependent Claim
60 words
The computer-readable medium according to claim 39, wherein for the first common item the test statistic corresponds to a probability that a third user will have a range of randomized transformed ratings that is at least as close to the first transformed ratings data corresponding to the first user as the randomized transformed ratings data corresponding to the second user.
Claim 46
Dependent Claim
14 words
The computer-readable medium according to claim 41, wherein the third user is randomly selected.
Claim 47
Dependent Claim
33 words
The computer-readable medium according to claim 39, further including computer readable code for determining at least one similarity value for the first user with respect to second ones of the plurality of users.
Claim 48
Dependent Claim
46 words
The computer-readable medium according to claim 43, further including computer readable code for determining a subgroup of users for recommending the first item based on the at least one similarity value of the first user with respect to the second ones of the plurality of users.
Claim 49
Dependent Claim
23 words
The computer-readable medium according to claim 39, wherein the computer readable code further includes code for computing the similarity value using a z-score.
Claim 50
Dependent Claim
24 words
The computer-readable medium according to claim 39, wherein the computer readable code further includes code for computing the similarity value using a Chi-square distribution.
Claim 51
Dependent Claim
42 words
The computer-readable medium according to claim 39, wherein the computer readable code further includes code for determining the similarity value from a plurality of test statistics each corresponding to further items that the first and second users have provided ratings data for.
Claim 52
Dependent Claim
44 words
The computer-readable medium according to claim 39, wherein the computer readable code further includes code for computing the range of randomized ratings data by adding a uniformly distributed random number to the ratings data provided by the first ones of the plurality of users.
Claim 53
Dependent Claim
42 words
The computer-readable medium according to claim 39, further including computer readable code for determining further common items that the first and second users have provided ratings for, and assigning a statistical weight to the first common item and the further common items.
Claim 54
Dependent Claim
30 words
The computer-readable medium according to claim 49, wherein the computer readable code for assigning a statistical weight further includes code for assigning a statistical weight using a genetic algorithm technique.
Claim 55
Dependent Claim
25 words
The computer-readable medium according to claim 40, further including computer readable code for providing a review of the first recommended item to the first user.
Claim 56
Dependent Claim
48 words
The computer-readable medium according to claim 40, further including compute readable code for providing a plurality of reviews of the first recommended item by further users to the first user in an order corresponding to respective similarity values of the first user with respect to the further users.
Claim 57
Dependent Claim
29 words
The computer-readable medium according to claim 40, further including computer readable code for allowing the first user to rate reviews of the first recommended item provided by further users.
Claim 58
Dependent Claim
40 words
The computer-readable medium according to claim 53, further including computer readable code for determining a goodness value for at least some of the reviews of the recommended item from the ratings given to the respective review by the first user.
Claim 59
Dependent Claim
29 words
The computer-readable medium according to claim 54, further including compute readable code for ordering the reviews provided to the first user using the goodness value associated with each review.
Claim 60
Dependent Claim
69 words
The computer-readable medium according to claim 39, further including computer readable code for evaluating a recommendation of an item wherein evaluating the recommendation includes recommending a second item that the first and second users have provided ratings for while ignoring the ratings data of the second recommended item provided by the first user, and examining the actual ratings data of the second recommended item provided by the first user.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2002-03-15 | 2002-09-14 | 2003-03-16 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2006-03-15 | 2006-09-14 | 2007-03-16 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2010-03-15 | 2010-09-14 | 2011-03-16 | $7700.00 | Window Closed / Lapsed |