Fraud detection using text analysis
View Patent ↗In one embodiment, a method executed by at least one processor includes receiving text from submitted by a user. The method also includes determining a text score for the received text by comparing a first set of phrases included in the received text to a second set of phrases. The second set of phrases includes phrases from stored text. The stored text includes stored text known to be genuine and stored text known to be fraudulent. The method also includes determining that the received text is fraudulent based on the text score.
1. A method executed by at least one processor comprising the steps of:
receiving, at a system, a profile submitted by a user, the system configured to compare a set of profiles to one another;
determining a geographical location from which the profile was submitted;
determining that the geographical location from which the profile was submitted is a geographical location associated with fraudulent profile submissions;
determining a text score for the profile by comparing a first set of phrases included in the profile to a second set of phrases, the second set of phrases comprising phrases from stored text, the stored text comprising stored text known to be genuine and stored text known to be fraudulent;
determining that the profile is fraudulent using the text score and the geographical location from which the profile was submitted; and
in response to determining that the profile is fraudulent, preventing the system from comparing the profile to the set of profiles.
2. The method of claim 1 , wherein determining the text score for the profile by comparing the first set of phrases to the second set of phrases further comprises:
determining a phrase score for each phrase of the first set of phrases by comparing the words of each phrase of the first set of phrases to the second set of phrases; and
determining the text score by adding the phrase scores of each phrase of the first set of phrases.
3. The method of claim 1 , wherein the stored text is organized in at least one tree structure.
4. The method of claim 1 , wherein determining that the profile is fraudulent comprises comparing the text score to a threshold.
5. The method of claim 4 , further comprising determining a threshold using a characteristic of the user and wherein comparing the text score to the threshold comprises comparing the text score to the determined threshold.
6. The method of claim 1 , wherein comparing the first set of phrases included in the profile to a second set of phrases comprises generating a tree structure corresponding to the first set of phrases.
7. The method of claim 1 , wherein the profile is determined to be fraudulent using the text score and using submissions by the user received before the profile.
8. The method of claim 1 , wherein the geographical location comprises a geographical region.
9. The method of claim 4 , further comprising determining the threshold using the deteimination that the profile was submitted from a geographical location associated with fraudulent profile submissions.
10. A system comprising:
an interface configured to receive a profile submitted by a user, the system configured to compare a set of profiles to one another; and
at least one processor configured to:
determine a geographical location from which the profile was submitted;
determine that the geographical location from which the profile was submitted is a geographical location associated with fraudulent submissions;
determine a text score for the profile by comparing a first set of phrases included in the profile to a second set of phrases, the second set of phrases comprising phrases from stored text, the stored text comprising stored text known to be genuine and stored text known to be fraudulent;
determine that the received text is fraudulent using the text score and the geographical location from which the profile was submitted; and
in response to determining that the profile is fraudulent, preventing the system from comparing the profile to the set of profiles.
11. The system of claim 10 , wherein the at least one processor configured to determine the text score for the profile by comparing the first set of phrases to the second set of phrases comprises the at least one processor configured to:
determine a phrase score for each phrase of the first set of phrases by comparing the words of each phrase of the first set of phrases to the second set of phrases; and
determine the text score by adding the phrase scores of each phrase of the first set of phrases.
12. The system of claim 10 , wherein the stored text is organized in at least one tree structure.
13. The system of claim 10 , wherein the at least one processor is configured to determine that the profile is fraudulent at least by comparing the text score to a threshold.
14. The system of claim 13 , wherein:
the at least one processor is further configured to determine a threshold using a characteristic of the user; and
the at least one processor configured to compare the text score to the threshold comprises the at least one processor configured to compare the text score to the determined threshold.
15. The system of claim 10 , wherein the at least one processor configured to compare the first set of phrases included in the profile to a second set of phrases comprises the at least one processor configured to generate a tree structure corresponding to the first set of phrases.
16. The system of claim 10 , wherein the at least one processor is further configured to determine that the profile is fraudulent using submissions by the user received before the profile.
17. The system of claim 10 , wherein the geographical location comprises a geographical region.
18. The system of claim 13 , wherein the at least one processor is further configured to determine the threshold using the determination that the profile was submitted from a geographical location associated with fraudulent profile submissions.
19. At least one non-transitory computer-readable medium comprising a plurality of instructions that, when executed by at least one processor, are configured to:
receive, at a system, a profile submitted by a user, the system configured to compare a set of profiles to one another;
determine a geographical location from which the profile was submitted;
determine that the geographical location from which the profile was submitted is a geographical location associated with fraudulent profile submissions;
determine a text score for the profile by comparing a first set of phrases included in the profile to a second set of phrases, the second set of phrases comprising phrases from stored text, the stored text comprising stored text known to be genuine and stored text known to be fraudulent;
determine that the profile is fraudulent using the text score and the geographical location from which the profile was submitted; and
in response to determining that the profile is fraudulent, preventing the system from comparing the profile to the set of profiles.
20. The at least one non-transitory computer-readable medium of claim 19 , wherein the plurality of instructions configured to determine the text score for the profile comprises the plurality of instructions configured to:
determine a phrase score for each phrase of the first set of phrases by comparing the words of each phrase of the first set of phrases to the second set of phrases; and
determine the text score by adding the phrase scores of each phrase of the first set of phrases.
21. The at least one non-transitory computer-readable medium of claim 9 , wherein the stored text is organized in at least one tree structure.
22. The at least one non-transitory computer-readable medium of claim 19 , wherein the plurality of instructions are configured to determine that the profile is fraudulent at least by comparing the text score to a threshold.
23. The at least one non-transitory computer-readable medium of claim 22 , wherein:
the plurality of instructions are further configured to determine a threshold using a characteristic of the user; and
the plurality of instructions configured to compare the text score to the threshold comprises the plurality of instructions configured to compare the text score to the determined threshold.
24. The at least one non-transitory computer-readable medium of claim 19 , wherein the plurality of instructions configured to compare the first set of phrases included in the profile to a second set of phrases comprises the plurality of instructions configured to generate a tree structure corresponding to the first set of phrases.
25. The at least one non-transitory computer-readable medium of claim 19 , wherein the plurality of instructions are further configured to determine that the profile is fraudulent using the text score and using submissions by the user received before the profile.
26. The at least one non-transitory computer-readable medium of claim 19 , wherein the geographical location comprises a geographical region.
27. The at least one non-transitory computer-readable medium of claim 22 , wherein the plurality of instructions are further configured to determine the threshold using the determination that the profile was submitted from a geographical location associated with fraudulent profile submissions.