System and method for dynamic profiling of users in one-to-one applications and for validating user rules

• US 6,871,186 B1
• Filed: 11/13/1998
• Issued: 03/22/2005
• Est. Priority Date: 11/14/1997
• Status: Expired due to Term

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: New York University
• Inventors: Tuzhilin, Alexander S., Adomavicius, Gediminas
• Primary Examiner(s): Smith, Jeffrey A.
• Assistant Examiner(s): Garg, Yogesh C
• Application Number:
  US09/554,383
• Time in Patent Office: 2,321 Days
• Field of Search: 705/26, 705/1, 705/14, 347/721, 379/189, 707/3, 707/1, 707/104, 707/2, 707/6, 706/11, 706/16, 706/20, 706/12, 706/45, 706/52, 709/217, 345/721
• US Class Current: 705/26.7
• CPC Class Codes: G06Q 30/018 Certifying business or products
• 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
• CPC Class Codes: G06Q 30/0224 based on user history
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 30/0267 Wireless devices
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: Y10S 706/925 Business
• CPC Class Codes: Y10S 706/934 Information retrieval or Information management
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A system and method for generating and validating a user profile (25) for a user based on a static profile (10) and a dynamic profile (15) of the user. The method and system compresses the dynamic rules (15) into aggregated rules so that the user can view a comparatively small number of the aggregated rules and select the desired rules from the aggregated rules based on user-desired criteria. The method and system validates user rules (60) using a processing device, which are retrieved from a storage device. The user rules are seperated into at least one subset of a user set. Then, it is determined if a particular rule of the at least one subset is one of acceptable, unacceptable and undecided based on a defined criteria (415). If the particular rules of the at least one subset are acceptable, the particular rules of the at least one subset are provided (e.g. assigned) to a corresponding user (435).

Claims

1. A method for generating a user profile for a user based on a static a profile and a dynamic profile of the user, the static profile including factual user information, the dynamic profile including user dynamic rules provided as a function of transactional user information, the method comprising the steps of:

(a) receiving the factual user information and the user dynamic rules;

(b) generating the static profile as a function of the factual and transactional user information;

(c) using a processing device for compressing the user dynamic rules into user aggregated rules;

(d) providing the user aggregated rules to at least one of the user and an expert;

e) selecting by at least one of the user and the expert, at least one aggregated rule from the aggregated rules;

f) matching the user dynamic rules to the at least one selected aggregated rule to generate the dynamic profile; and

g) combining the static profiles and the dynamic profile to form the particular profile.

2. The method according to claim 1, wherein step (e) includes the substep of:

h) validating the user aggregated rules in the dynamic profile.

3. The method according to claim 1, wherein step (c) includes the following substeps:

I. determining a plurality of similar dynamic rules from the user dynamic rules,

ii. combining the plurality of similar dynamic rules into at least one corresponding cluster, and

iii. generating the user aggregated rules as a function of the at least one corresponding cluster.

4. The method according to claim 3, wherein the at least one cluster includes a plurality of clusters, and wherein substep (iii) includes the following substeps:

A) determining a first representative rule for each cluster of the plurality of clusters, and

B) if a number of the plurality of clusters is greater that a predetermined threshold number, compressing the plurality of clusters into a smaller number of the plurality of clusters.

5. The method according to claim 4, wherein substep (iii) further includes the following substeps:

C) identifying users providing the first representative rule which corresponds to a particular cluster of the plurality of clusters to form a user cluster,

D) determining a second representative expression for the user cluster,

E) augmenting the first representative rules and the second representative rules to form combined rules, and

F) converting the combined rules into the user aggregated rules.

6. The method according to claim 3, wherein each of the user aggregated rules is determined by obtaining a center of the at least one corresponding cluster.

7. The method according to claim 5, wherein the user aggregated rules include fuzzy logic characteristics.

8. The method according to claim 4, wherein substep (B) includes the following substeps:

I. selecting a first cluster and a second cluster from the plurality of clusters,

II. determining a cluster distance between the first cluster and the second cluster,

III. determining a first size of the first cluster and a second size of the second cluster,

IV. if the cluster distance between the first cluster and the second cluster is smaller than or equal to a predetermined relation between the first size and the second size, merging the first cluster and the second cluster to form a merged cluster, and

V. if the cluster distance is larger than the predetermined relation, selecting a further first cluster and a further second cluster and repeating substeps II through IV using the further first cluster as the first cluster and using the further second cluster as the second cluster.

9. The method according to claim 8, wherein substep (B) includes the following substep:

VI. determining a center cluster of the plurality of clusters from the merged cluster.

10. The method according to claim 1, wherein step (a) further includes a step of retrieving a previous dynamic profile of the user, and wherein step (g) includes a step of combining the previous dynamic profile to the static profile and the dynamic profile to form the user profile.

11. A method for providing suggestions to a user based on a user profile associated with the user, comprising the steps of:

a) receiving user current state information associated with the user;

b) retrieving a static profile associated with the user, the static profile including factual user information corresponding to user preferences;

c) retrieving a dynamic profile associated with the user, the dynamic profile including rules corresponding to user repetitive transactions;

d) using a processing device for compressing the rules into user of the dynamic profile to form aggregated rules as a function of a predetermined similarity criteria;

e) providing the aggregated rules corresponding to the dynamic profile to an expert;

f) selecting at least one of rule from the aggregated rules based on a user desired criteria;

g) matching user dynamic rules to the at least one selected rule to update the dynamic profile; and

h) combining the static profile and the dynamic profile to form the user profile;

i) providing recommendation to the user as a function of the user profile and the user current state information.

12. The method according to claim 11, further comprising the step of:

j) after step (h) and before step (i), receiving present state-of-the-world information corresponding to the user profile and the user current state information,

wherein step (i) further includes a step of providing recommendations as a further function of the present state-of-the-world information.

13. The method according to claim 12, further comprising the step of:

k) after step (a) and before step (b), receiving past transactional information associated with the user.

14. The method according to claim 12, wherein step (d) includes the following substeps:

I. determining a plurality of similar dynamic rules from the user dynamic rules,

ii. combining the plurality of similar dynamic rules into at least one corresponding cluster, and

iii. generating the user aggregated rules as a function of the at least one corresponding cluster.

15. A system for generating a user profile for a user based on a static profile and a dynamic profile of the user, the static profile including factual user information, the dynamic profile including user dynamic rules corresponding to transactional user information, the system comprising:

a storage arrangement storing the static and dynamic profiles of the user; and

a processor receiving the factual user information and the user dynamic rules from the storage arrangement and generating the static and dynamic profile as a function of the factual user information, the user dynamic rules being compressed by the processor to form user aggregated rules,

wherein the processor provides the user aggregated rules to at least one of the user and an expert,

wherein the processor receives data from at least one of the user and the expert to select at least one aggregated rule of the aggregated rules

wherein the processor matches the user dynamic rules to the at least one selected aggregated rule to generate the dynamic profile, and

wherein the static profile and the dynamic profile are combined to form the user profile.

16. A method for providing a recommendation to a user based on a user profile associated with the user, wherein the user profile includes a dynamic profile including user dynamic rules, comprising the steps of,

(a) receiving user current state information and transactional information associated with the user,

(b) storing the transactional information as at least a portion of a user purchasing history;

(c) providing the user purchasing history to a user profile generation module and generating the user profile as a function of the user purchasing history;

(d) using a processing device for compressing the user dynamic rules into user aggregated rules so as to improve the user profile

(e) estimating user needs information as a function of the user profile; and

(f) generating the recommendation to the user as a function of the user estimated needs information and state-of-the world information and providing the recommendation to the user using a remote unit.

17. A system for providing a recommendation to a user based on a user profile associated with the user, wherein the user Profile includes a static profile and a dynamic profile including user dynamic rules, comprising:

a first module receiving user current state information and transactional information associated with the user;

a storage device storing as at least a portion of a user purchasing history obtained from the transactional information;

a second module receiving the user purchasing history and generating the user profile as a function of the user purchasing history;

using a processing device for compressing the user dynamic rules into user aggregated rules so as to improve the user profile;

a third module estimating the user needs information as a function of the user profile;

a fourth module generating the recommendation to the user as a function of the user estimated needs information and state-of-the world information; and

an output device providing the recommendation to the user using a remote unit.

18. A method for generating a particular profile for a user based on a static profile and a dynamic profile of the user, the static profile including factual information for the user, the dynamic profile including dynamic rules provided as a function of transactional information for the user, the method comprising the steps of:

a) receiving the factual information and the dynamic rules;

b) generating the static profile as a function of the factual and transactional information;

c) using a processing device for compressing the dynamic rules into aggregated rules;

d) providing the aggregated rules to an expert;

e) selecting, by the expert, at least one aggregated rule from the aggregated rules;

f) matching the dynamic rules to the at least one selected aggregated rule to generate the dynamic profile; and

g) combining the static profile and the dynamic profile to form the particular profile.

19. A system for generating a particular profile for a user based on a static profile and a dynamic profile of the user, the static profile including factual information corresponding to substantially fixed information for the user, the dynamic profile including dynamic rules corresponding to transactional information for the user, the system comprising:

a storage arrangement storing the static and dynamic profiles of the user; and

a processor receiving the factual information and the dynamic rules from the storage arrangement, the processor generating the static and dynamic profile as a function of the factual information, the dynamic rules being compressed by the processor to form aggregated rules,

wherein the processor provides the aggregated rules to an expert, and receives data from the expert to select at least one aggregated rule of the aggregated rules based on a desired criteria,

wherein the processor matches the dynamic rules to the at least one selected aggregated rule to generate the dynamic profile, and

wherein the processor combines the static profile and the dynamic profile to form the particular profile.

20. A method for providing first information to a user based on a particular profile associated with the user, comprising the steps of:

a) receiving current state information and transactional information associated with the user;

b) storing the transactional information as at least a portion of a transaction history for the user;

c) providing the transaction history to a profile generation module for;

generating the particular profile as a function of the transaction history, the particular profile including a static profile and a dynamic profile, the static profile including factual information, the dynamic profile including dynamic rules provided as a function of the transactional information for the user;

d) compressing the dynamic rules into aggregated rules to improve the dynamic profile;

e) estimating second information for the user as a function of the transaction history, the current state information of the user and the particular profile; and

f) providing the first information to the user as a function of the estimated second information.

21. A system for providing first information to a user based on a particular profile associated with the user, comprising:

a first module receiving current state information and transactional information associated with the user;

a storage device storing the transactional information as at least a portion of a transaction history associated with the user;

a second module for;

receiving the transaction history and generating the particular profile as a function of the transaction history, the particular profile including a static profile and a dynamic profile, the static profile including factual information, the dynamic profile including dynamic rules provided as a function of the transactional information for the user; and

compressing the dynamic rules into aggregated rules to improve the dynamic profile;

a third module estimating second information of the user as a function of needs information for the user as a function of the transaction history, current state information of the user and the particular profile; and

a fourth module providing the first information to the user as a function of the estimated second information.