Methods, devices and systems for data augmentation to improve fraud detection

• US 10,664,656 B2
• Filed: 06/20/2018
• Issued: 05/26/2020
• Est. Priority Date: 06/20/2018
• Status: Active Grant

• Current Assignee: VADE USA Incorporated
• Original Assignee: VADE Secure Incorporated
• Inventors: Goutal, Sebastien, Meyer, Maxime Marc
• Primary Examiner(s): Dyer, Andrew R
• Application Number:
  US16/013,581
• Publication Number:
  US 20190392038A1
• Time in Patent Office: 706 Days
• Field of Search: 715255, 715256, 715264, 715271
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence
• CPC Class Codes: G06F 21/30 Authentication, i.e. establishing the identity or authorisation of security principals
• CPC Class Codes: G06F 21/64 Protecting data integrity, e.g. using checksums, certificates or signatures
• CPC Class Codes: G06F 40/157 using dictionaries or tables
• CPC Class Codes: G06F 40/166 Editing, e.g. inserting or deleting
• CPC Class Codes: G06F 40/194 Calculation of difference between files
• CPC Class Codes: G06F 40/216 using statistical methods
• CPC Class Codes: G06F 40/232 Orthographic correction, e.g. spell checking or vowelisation
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06N 3/045 Combinations of networks
• CPC Class Codes: G06N 3/08 Learning methods
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A computer-implemented method of generating an augmented electronic text document comprises establishing a directed multigraph where each vertex is associated with a separate language and is connected to at least one other one of the vertices by an oriented edge indicative of a machine translation engine'"'"'s ability to translate between languages associated with the vertices connected by the oriented edge with acceptable performance. The directed multigraph is then traversed starting at a predetermined origin vertex associated with an original language of the original electronic text document by randomly selecting an adjacent vertex pointed to by an oriented edge connected to the predetermined origin vertex and causing a machine translation engine to translate the original electronic text document from the original language to a language associated with the selected vertex. The directed multigraph is then further traversed as allowed by the oriented edges from the intermediate vertex to successive other next-adjacent connected vertices, each time machine translating a previously-translated electronic text document into a language associated with a randomly-selected next-adjacent vertex until the predetermined origin vertex is selected and the previously translated electronic text document is re-translated into the original language and designated as the augmented electronic text document.

Claims

1. A computer-implemented method for modifying an original electronic text document of a corpus of electronic text documents, comprising:

receiving the original electronic text document in a computer having a memory;

repeatedly translating the received original electronic text document, using at least one machine translation engine, wherein each translated electronic text document is used as a basis for a subsequent translation into another language;

re-translating a last-translated electronic text document back into an original language of the original electronic text document;

transforming the re-translated electronic text document by selecting at least one word therein and substituting a respective synonym for each selected word to generate a synonym-replaced electronic text document;

transforming the synonym-replaced electronic text document by selecting at least one word therein and substituting a respective misspelled word for each selected word to generate an modified electronic text document;

computing a similarity measure between the original electronic text document and the modified electronic text document;

determining whether the computed similarity measure is at least as great as a predetermined similarity threshold; and

storing the modified electronic text document in the memory if the computed similarity threshold is greater than or equal to the predetermined similarity threshold and not storing the modified electronic text document in the memory if the computed similarity threshold is less than the predetermined similarity threshold.

2. The computer-implemented method of claim 1, wherein repeatedly translating, transforming the re-translated electronic text document and transforming the synonym-replaced electronic text document are performed sequentially, in order.

3. The computer-implemented method of claim 1, wherein each of the translating and transforming steps is configured to be sequentially performed a predetermined number of times.

4. The computer-implemented method of claim 1, further comprising selecting at least one word in the re-translated or synonym-replaced electronic text document and replacing the selected at least one word with a respective abbreviation.

5. The computer-implemented method of claim 1, further comprising selecting at least one entity in the re-translated or synonym-replaced electronic text document and replacing the selected at least one entity with a replacement entity.

6. The computer-implemented method of claim 5, wherein the selected at least one entity comprises at least one of a first name of an individual, last name of an individual, a financial amount, a date and a telephone number.

7. The computer-implemented method of claim 1, wherein computing the similarity measure comprises computing at least a semantic similarity measure between the original electronic text document and the modified electronic text document.

8. The computer-implemented method of claim 1, wherein computing the similarity measure comprises computing at least a syntactical similarity measure between the original electronic text document and the modified electronic text document.

9. A computing device comprising:

at least one processor;

at least one data storage device coupled to the at least one processor;

a network interface coupled to the at least one processor and to a computer network;

a plurality of processes spawned by the at least one processor to modify an original electronic text document of a corpus of electronic text documents, the processes including processing logic for;

repeatedly translating the original electronic text document, using at least one machine translation engine, wherein each translated text document is used as a basis for a subsequent translation into another language;

re-translating a last-translated electronic text document back into an original language of the original electronic text document;

transforming the re-translated electronic text document by selecting at least one word therein and substituting a respective synonym for each selected word to generate a synonym-replaced electronic text document;

transforming the synonym-replaced electronic text document by selecting at least one word therein and substituting a respective misspelled word for each selected word to generate an modified electronic text document;

computing a similarity measure between the original electronic text document and the modified electronic text document;

determining whether the computed similarity measure is at least as great as a predetermined similarity threshold; and

storing the modified text in the data storage device if the computed similarity measure is greater than or equal to the predetermined similarity threshold and discarding and not storing the modified electronic text document in the data storage device if the computed similarity measure is less than the predetermined similarity threshold.

10. The computing device of claim 9, wherein the processing logic for repeatedly translating, transforming the re-translated text and transforming the synonym-replaced electronic text document is performed sequentially, in order.

11. The computing device of claim 9, wherein the processing logic for each of the translating and transforming steps is configured to be sequentially executed a predetermined number of times.

12. The computing device of claim 9, further comprising processing logic for selecting at least one word in the re-translated or synonym-replaced electronic text document and replacing the selected at least one word with a respective abbreviation.

13. The computing device of claim 9, further comprising processing logic for selecting at least one entity in the re-translated or synonym-replaced electronic text document and for replacing the selected at least one entity with a replacement entity.

14. The computing device of claim 13, wherein the selected at least one entity comprises at least one of a first name of an individual, last name of an individual, a financial amount, a date and a telephone number.

15. The computing device of claim 9, wherein the processing logic for computing the similarity measure comprises processing logic for computing at least a semantic similarity measure between the original electronic text document and the modified electronic text document.

16. The computing device of claim 9, wherein the processing logic for computing the similarity measure comprises processing logic for computing at least a syntactical similarity measure between the original electronic text document and the modified electronic text document.