Granted Patent B2

US 12,373,552 · App. 18/622,553 · Granted Jul 29, 2025

Inventors: Thomas Levi (Vancouver, CA); Steve Oldridge (Vancouver, CA)

Assignee: PLENTYOFFISH MEDIA ULC

G06F21/552G06F21/577

Abstract

Claims (245)

1. A method of detecting suspected fraudulently generated user identity profiles in an online network environment using a fraud detection system that includes at least one processor and at least one non-transitory processor-readable medium that stores processor-executable instructions, the online network environment including an attribute value for an attribute associated with the user identity profiles, and including an additional attribute value that occurs frequently in the set of profiles, the method comprising:

determining, via the at least one processor, whether the attribute value is linked to a high conditional probability based on the attribute value that a profile is fraudulently generated or undesirable;

identifying, via the at least one processor, a set of profiles likely to be fraudulently generated or undesirable based on the attribute value and the high conditional probability;

querying, using the fraud detection system having at least one processor and at least one non-transitory processor-readable medium that stores processor-executable instructions, a data store containing profile information to identify profiles suspected of being fraudulently generated based at least in part on the attribute value and the additional attribute value;

screening, using the fraud detection system having at least one processor and at least one non-transitory processor-readable medium that stores processor-executable instructions, the identified profiles suspected of being fraudulently generated or undesirable; and

quarantining the identified profiles.

2. The method of claim 1 , wherein determining whether the attribute value is linked to a high conditional probability of being a fraudulent or undesirable profile includes computing the initial probabilities of a profile being a fraudulent profile [p(S)] or a valid profile [p(V)].

3. The method of claim 1 , wherein determining whether the attribute value is linked to a high conditional probability includes computing the value indicative of a conditional probability that the respective profile is fraudulent according to:

p

⁡

(

S

❘

{

x

i

}

)

=

p

⁡

(

S

)

⁢

∏

i

M

p

⁡

(

x

i

❘

S

)

p

⁡

(

S

)

⁢

∏

i

M

p

⁡

(

x

i

❘

S

)

+

p

⁡

(

V

)

⁢

∏

i

M

p

⁡

(

x

i

❘

V

)

.

4. The method of claim 1 , wherein the attribute value includes one or more of:

a Hyper Text Transfer Protocol (http) referrer associated with the respective profile;

an Internet Protocol (IP) of signup and last logins associated with the respective profile;

one or more cookies used to track individual computers associated with the respective profile;

one or more cookies that contain a user identifier of the most recent users to log in on using a given instance of a processor-based device; and

IP blocks of signup and at least two most recent logins associated with the respective profile.

5. The method of claim 1 , further comprising:

ranking the profiles into groups.

6. The method of claim 1 , further comprising: providing the designated profiles to a front end system for possible deletion.

7. A user identity profile detection system in an online network environment to detect at least one of accounts or related profiles suspected of being undesirable, the online network environment including an initial probability value based at least in part on historical profile data, and including an additional attribute value that occurs frequently in the set of profiles, the system comprising:

at least one processor; and

at least one non-transitory processor-readable medium that stores processor-executable instructions, wherein the at least one processor:

determine whether the initial probability value is linked to a value indicative of a high conditional probability based on the initial probability value that a profile is undesirable;

identify a set of profiles likely to be undesirable based on the initial probability value and the high conditional probability;

query a data store containing profile information to identify profiles suspected of being undesirable based at least in part on the initial probability value and the additional attribute value;

screen the identified profiles suspected of being undesirable; and

quarantine the identified profiles.

8. The system of claim 7 , wherein the at least one processor computes a value indicative of the initial probabilities of a profiles being at least one of an undesirable-profile [p(S)] or a valid profile [p(V)].

9. The system of claim 7 , wherein the at least one processor computes a value indicative of a conditional probability that the respective profile is undesirable includes computing the value indicative of a conditional probability that the respective profile is undesirable according to:

p

⁡

(

S

❘

{

x

i

}

)

=

p

⁡

(

S

)

⁢

∏

i

M

p

⁡

(

x

i

❘

S

)

p

⁡

(

S

)

⁢

∏

i

M

p

⁡

(

x

i

❘

S

)

+

p

⁡

(

V

)

⁢

∏

i

M

p

⁡

(

x

i

❘

V

)

.

10. The system of claim 7 , wherein the at least one processor ranks the profiles into groups using the high conditional probability based on the initial probability value that the profile is undesirable.

11. The system of claim 7 , wherein the attribute value includes one or more of:

a Hyper Text Transfer Protocol (http) referrer associated with the respective profile;

an Internet Protocol (IP) of signup and last logins associated with the respective profile;

one or more cookies used to track individual computers associated with the respective profile;

one or more cookies that contain a user identifier of the most recent users to log in on using a given instance of a processor-based device; and

IP blocks of signup and at least two most recent logins associated with the respective profile.

12. The system of claim 7 , wherein the at least one processor generates at least one output logically associated with each profile, the at least one output indicative of at least one of the following: a deletion indicator, a clearance indicator, or a further investigation indicator.

13. A fraudulent user identity profile detection system in an online network environment to detect at least one of accounts or related profiles suspected of being fraudulently generated, the online network environment including an attribute value for an attribute associated with the user identity profiles, and including an additional attribute value that occurs frequently in the set of profiles, the system comprising:

at least one processor; and

at least one nontransitory processor-readable medium that stores processor-executable instructions, wherein the at least one processor:

determine whether the attribute value is linked to a high conditional probability based on the attribute value that a profile is fraudulently generated;

identify a set of profiles likely to be fraudulently generated based on the attribute value and the conditional probability;

query a data store containing profile information to identify profiles suspected of being fraudulently generated based at least in part on the attribute value and the additional attribute value;

screen the identified profiles suspected of being fraudulently generated; and

quarantine the identified profiles.

14. The system of claim 13 , wherein the at least one processor computes a value indicative of an initial probabilities of a profiles being at least one of a fraudulent profile [p(S)] or a valid profile [p(V)].

15. The system of claim 13 , wherein the at least one processor computes a value indicative of a conditional probability that the respective profile is fraudulent includes computing the value indicative of a conditional probability that the respective profile is fraudulent according to:

p

⁡

(

S

❘

{

x

i

}

)

=

p

⁡

(

S

)

⁢

∏

i

M

p

⁡

(

x

i

❘

S

)

p

⁡

(

S

)

⁢

∏

i

M

p

⁡

(

x

i

❘

S

)

+

p

⁡

(

V

)

⁢

∏

i

M

p

⁡

(

x

i

❘

V

)

.

16. The system of claim 13 , wherein the at least one processor ranks the profiles into groups using the high conditional probability based on the initial probability value that the profile is one of either fraudulent.

17. The system of claim 13 , wherein the attribute value includes one or more of:

a Hyper Text Transfer Protocol (http) referrer associated with the respective profile;

an Internet Protocol (IP) of signup and last logins associated with the respective profile;

one or more cookies used to track individual computers associated with the respective profile;

one or more cookies that contain a user identifier of the most recent users to log in on using a given instance of a processor-based device; and

IP blocks of signup and at least two most recent logins associated with the respective profile.

18. The system of claim 13 , wherein the at least one processor generates at least one output logically associated with each profile, the at least one output indicative of at least one of the following: a deletion indicator, a clearance indicator, or a further investigation indicator.

Continuity (6)

References Cited (317)