System and method for identifying abusive account registration
Disclosed is a system and method for processing account registration by identifying account candidates attempting to open an account as abusive. That is, the present disclosure discusses identifying, and challenging and marking abusive account registration. The present disclosure takes into account users' behaviors on a network and the impact to the cost and/or revenue of the network. The present disclosure is proactive as it allows for actions to be taken at the earliest possible time in the registration process before an account is created. This prevents abusive activity from taking place within the network and effecting services and privileges available to legitimate users. Additionally, the effects of the disclosed systems and methods minimize the negative impacts of abusive activity on normal user accounts.
1. A method comprising:
collecting, via a computing device, user activity data associated with an existing user account on a network;
identifying, via the computing device, a pattern within said user activity data;
determining, via the computing device, whether said pattern represents legitimate or abusive activity;
training, via the computing device, a classifier to analyze a new account registration request based upon said pattern determination, said training comprising providing the classifier with activity information related to legitimate and abusive activity, said activity information is based upon said pattern determination; and
applying, via the computing device, the classifier to the new account registration request in order to determine whether the new account registration is abusive.
2. The method of claim 1 , wherein said applying the classifier comprises:
receiving, over the network, the new account registration request comprising user provided registration information;
extracting features of the user provided registration information;
comparing the features of the user provided registration information with the activity information; and
determining whether the request for account creation should be marked as abusive based upon said comparison.
3. The method of claim 2 , wherein if the features of the user provided registration information match said activity information related to abusive activity, the request is challenged.
4. The method of claim 3 , further comprising:
identifying the user provided registration information as abusive; and
providing the identified user provided registration information to the classifier for training.
5. The method of claim 2 , wherein if the user provided registration information matches said activity information related to legitimate activity, the request is marked legitimate.
6. The method of claim 1 , further comprising:
identifying the user account associated with the user activity data as legitimate or abusive based upon said pattern determination;
labeling the user account as legitimate or abusive in accordance with said identification; and
providing information related to said labeling to the classifier for training.
7. The method of claim 6 , wherein labeling the user account comprises:
labeling the user activity data associated with the user account as legitimate or abusive based upon said pattern determination.
8. The method of claim 6 , wherein said labeling the user account further comprises:
labeling account registration information associated with the labeled user account in accordance with how the user account is labeled.
9. The method of claim 6 , further comprising:
updating an account database with said information related to said labeling the user account, said account database comprising user account information for user accounts on the network.
10. The method of claim 1 , wherein the classifier implements machine learning techniques in accordance with the activity information.
11. The method of claim 1 , wherein applying the classifier further comprises:
determining a probability score of abusive activity for the new account registration request based upon said activity information; and
comparing the probability score against a threshold,
wherein, if the probability score is equal to or greater than said threshold, marking the new account registration as abusive, and
wherein, if the probability score is less than the threshold, marking the new account registration request as legitimate.
12. The method of claim 2 , wherein the classifier is trained and updated continuously in accordance with said determination respective of the request for account creation.
13. A computer-readable storage medium tangibly encoded with computer-executable instructions, that when executed by a computer, perform a method comprising:
collecting user activity data associated with an existing user account on a network;
identifying a pattern within said user activity data;
determining whether said pattern represents legitimate or abusive activity;
training a classifier to analyze a new account registration request based upon said pattern determination, said training comprising providing the classifier with activity information related to legitimate and abusive activity, said activity information is based upon said pattern determination; and
applying the classifier to the new account registration request in order to determine whether to mark the new account registration legitimate or abusive.
14. The computer-readable storage medium of claim 13 , wherein said applying the classifier comprises:
receiving, over the network, the new account registration request comprising user provided registration information;
extracting features of the user provided registration information;
comparing the features of the user provided registration information with the activity information; and
determining whether the request for account creation should be marked as abusive based upon said comparison,
wherein if the features of the user provided registration information match said activity information related to abusive activity, the request is marked abusive, further comprising:
identifying the user provided registration information as abusive; and
providing the identified user provided registration information to the classifier for training; and
wherein if the features of the user provided registration information match said activity information related to legitimate activity, the request is marked legitimate.
15. The computer-readable storage medium of claim 13 , further comprising:
identifying the user account associated with the user activity data as legitimate or abusive based upon said pattern determination;
labeling the user account as legitimate or abusive in accordance with said identification; and
providing information related to said labeling to the classifier for training.
16. The computer-readable storage medium of claim 15 , wherein labeling the user account comprises:
labeling the user activity data associated with the user account as legitimate or abusive based upon said pattern determination; and
labeling account registration information associated with the labeled user account in accordance with how the user account is labeled.
17. The computer-readable storage medium of claim 13 , wherein applying the classifier further comprises:
determining a probability score of abusive activity for the new account registration request based upon said activity information; and
comparing the probability score against a threshold,
wherein, if the probability score is equal to or greater than said threshold, marking the new account registration as abusive, and
wherein, if the probability score is less than the threshold, marking the new account registration as legitimate.
18. A system comprising:
at least one computing device comprising:
memory storing computer-executable instructions; and
one or more processors for executing said computer-executable instructions, comprising:
collecting user activity data associated with an existing user account on a network;
identifying a pattern within said user activity data;
determining whether said pattern represents legitimate or abusive activity;
training a classifier to analyze a new account registration request based upon said pattern determination, said training comprising providing the classifier with activity information related to legitimate and abusive activity, said activity information is based upon said pattern determination; and
applying the classifier to the new account registration request in order to determine whether to grant to deny account registration.
19. The system of claim 18 , wherein said applying the classifier comprises:
receiving, over the network, the new account registration request comprising user provided registration information;
comparing the user provided registration information with the activity information; and
determining whether the request for account creation should be marked legitimate based upon said comparison,
wherein if the user provided registration information matches said activity information related to abusive activity, the request is marked abusive, further comprising:
identifying the user provided registration information as abusive; and
providing the identified user provided registration information to the classifier for training; and
wherein if the user provided registration information matches said activity information related to legitimate activity, the request is marked legitimate.
20. The system of claim 18 , further comprising:
identifying the user account associated with the user activity data as legitimate or abusive based upon said pattern determination;
labeling the user account as legitimate or abusive in accordance with said identification; and
providing information related to said labeling to the classifier for training.