Apparatus and a method for detecting fraudulent signature inputs

1. An apparatus for detecting fraudulent signature inputs, wherein the apparatus comprises:

at least a processor; and

a memory communicatively connected to the at least a processor, wherein the memory containing instructions configuring the at least a processor to:

receive a plurality of image data from a user;

identify a plurality of signature elements as a function of the plurality of signature inputs;

determine a plurality of signature scores as a function of the plurality of signature elements, wherein the plurality of signature scores comprises a first set of signature scores and a second set of signature scores, wherein determining the plurality of signature elements additionally comprises:

iteratively training a score classifier using a plurality of score training data, wherein the plurality of score training data comprises the plurality of signature elements as inputs correlated to examples of signature scores as outputs; and

identifying the plurality of signature scores as a function of the plurality of image data using the trained score classifier;

generate an accuracy threshold as a function of the first set of signature scores; and

determine one or more fraudulent signature inputs from the plurality of signature inputs as a function of a comparison of the second set of signature scores to the accuracy threshold.

2. The apparatus of claim 1 , wherein the plurality of signature elements comprises at least a baseline orientation.

3. The apparatus of claim 1 , wherein the plurality of signature elements comprises at least a slant of a character.

4. The apparatus of claim 1 , wherein identifying the plurality of signature elements comprises identifying the plurality of signature elements using optical character recognition (OCR).

5. The apparatus of claim 1 , wherein identifying the plurality of signature elements comprises identifying the plurality of signature elements using a natural language processing model.

6. The apparatus of claim 1 , wherein identifying the plurality of signature elements additionally comprises:

identify one or more characters within each signature input of the plurality of signature inputs; and

identifying the plurality of signature elements associated with each character of the one or more characters.

7. The apparatus of claim 1 , wherein the memory further instructs the processor to identify one or more element clusters as a function of the first set of signature scores.

8. The apparatus of claim 7 , wherein generating the accuracy threshold comprises identifying a membership criterion of the one or more element clusters.

9. The apparatus of claim 8 , wherein comparing the second set of signature scores to the accuracy threshold comprises identifying whether the second set of signature scores satisfies the membership criteria for the one or more element clusters.

10. The apparatus of claim 1 , wherein the plurality of signature inputs comprises one or more historical signature inputs and one or more current signature inputs.

11. A method for detecting fraudulent signature inputs, wherein the method comprises:

receiving, using at least a processor, a plurality of image data from a user;

identifying, using the at least a processor, a plurality of signature elements as a function of the plurality of signature inputs;

determining, using the at least a processor, a plurality of signature scores as a function of the plurality of signature elements, wherein the plurality of signature scores comprises a first set of signature scores and a second set of signature scores, wherein determining the plurality of signature elements additionally comprises:

iteratively training a score classifier using a plurality of score training data, wherein the plurality of score training data comprises the plurality of signature elements as inputs correlated to examples of signature scores as outputs; and

identifying the plurality of signature scores as a function of the plurality of image data using a trained score classifier;

generating, using the at least a processor, an accuracy threshold as a function of the first set of signature scores; and

determining, using the at least a processor, one or more fraudulent signature inputs from the plurality of signature inputs as a function of a comparison of the second set of signature scores to the accuracy threshold.

12. The method of claim 11 , wherein the plurality of signature elements comprises at least a baseline orientation.

13. The method of claim 11 , wherein the plurality of signature elements comprises at least a slant of a character.

14. The method of claim 11 , wherein identifying the plurality of signature elements comprises identifying the plurality of signature elements using optical character recognition (OCR).

15. The method of claim 11 , wherein identifying the plurality of signature elements comprises identifying the plurality of signature elements using a natural language processing model.

16. The method of claim 11 , wherein identifying the plurality of signature elements additionally comprises:

identify one or more characters within each signature input of the plurality of signature inputs; and

identifying the plurality of signature elements associated with each character of the one or more characters.

17. The method of claim 11 , wherein the method further comprises identifying, using the at least a processor, one or more element clusters as a function of the first set of signature scores.

18. The method of claim 17 , wherein generating the accuracy threshold comprises identifying a membership criterion of the one or more element clusters.

19. The method of claim 18 , wherein comparing the second set of signature scores to the accuracy threshold comprises identifying whether the second set of signature scores satisfies the membership criteria for the one or more element clusters.

20. The method of claim 11 , wherein the plurality of signature inputs comprises one or more historical signature inputs and one or more current signature inputs.