Personal identifiable information encoder
A system can include a processor in communication with a data store. The processor can obtain personal identifiable information (PII) data and segregate the PII data into two or more secondary representations. The processor can generate a plurality of co-occurrence matrices based on the two or more secondary representations. The processor can perform a convolution between each of the plurality of co-occurrence matrices and one of a plurality of Gaussian kernels, wherein each of the plurality of Gaussian kernels comprises a different width. The processor can generate a tertiary representation of the PII data by performing a linear combination of the plurality of co-occurrence matrices. The processor can generate a vector based on the tertiary representation and perform a lossy tokenization process on the vector to generate a token. The processor can store the token at the data store.
1. An encoding system, comprising:
at least one processor in communication with at least one data store;
the at least one data store; and
a non-transitory, machine-readable memory device comprising instructions stored thereon that, when executed by the at least one processor, cause the at least one processor to:
obtain personal identifiable information (PII) data;
segregate the PII data into two or more secondary representations;
generate a plurality of co-occurrence matrices based on the two or more secondary representations;
perform a convolution between each of the plurality of co-occurrence matrices and one of a plurality of Gaussian kernels, wherein each of the plurality of Gaussian kernels comprises a different width;
generate a tertiary representation of the PII data by performing a linear combination of the plurality of co-occurrence matrices;
generate at least one vector based on the tertiary representation of the PII data;
perform a lossy tokenization process on the at least one vector to generate at least one token; and
store the at least one token at the data store.
2. The system of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to convert the PII data to a string format prior to segregating the PII data.
3. The system of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to:
compare the at least one token to at least one other token stored at the data store;
based on the comparison, determine that the at least one token is within a similarity threshold of the at least one other token; and
transmit, to a computing device, a positive verification of a subject associated with the at least one other token.
4. The system of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to, prior to generating the tertiary representation of the PII data, apply one of a plurality of weight values to each of the plurality of co-occurrence matrices.
5. The system of claim 4 , wherein:
individual values of the plurality of weight values are each associated with one of a plurality of categories; and
the instructions, when executed by the at least one processor, further cause the at least one processor to:
classify each of the plurality of co-occurrence matrices into one of the plurality of categories; and
apply the one of the plurality of weight values to each of the plurality of co-occurrence matrices by applying the individual value of the plurality of weight values corresponding to the one of the plurality of categories into which the co-occurrence matrix was classified.
6. The system of claim 5 , wherein:
the plurality of categories comprises a first name category, a last name category, and a middle name category; and
individual values of the plurality of weight values that are associated with first name category or the last name category are greater than individual values of the plurality of weight values associated with the middle name category.
7. The system of claim 4 , wherein the plurality of weight values are equal.
8. The system of claim 7 , wherein, prior to performing the convolution, the plurality of co-occurrence matrices are identical.
9. A non-transitory, computer-readable medium comprising instructions that, when executed by a computer, cause the computer to:
obtain personal identifiable information (PII) data;
segregate the PII data into two or more secondary representations;
generate a plurality of co-occurrence matrices based on the two or more secondary representations;
perform a convolution between each of the plurality of co-occurrence matrices and one of a plurality of Gaussian kernels, wherein each of the plurality of Gaussian kernels comprises a different width;
generate a tertiary representation of the PII data by performing a linear combination of the plurality of co-occurrence matrices;
generate at least one vector based on the tertiary representation of the PII data;
perform a lossy tokenization process on the at least one vector to generate at least one token; and
store the at least one token at the data store.
10. The non-transitory, computer-readable medium of claim 9 , wherein the instructions, when executed by the computer, cause the computer to:
generate scannable media based on the at least one token; and
transmit the scannable media to a computing device.
11. The non-transitory, computer-readable medium of claim 9 , wherein the instructions, when executed by the computer, cause the computer to convert the PII data to a string format prior to segregating the PII data.
12. The non-transitory, computer-readable medium of claim 9 , wherein the instructions, when executed by the computer, cause the computer to:
compare the at least one token to at least one other token stored at the data store;
based on the comparison, determine that the at least one token is within a similarity threshold of the at least one other token; and
transmit, to a computing device, a positive verification of a subject associated with the at least one other token.
13. The non-transitory, computer-readable medium of claim 9 , wherein the instructions, when executed by the computer, cause the computer to, prior to generating the tertiary representation of the PII data, apply one of a plurality of weight values to each of the plurality of co-occurrence matrices.
14. The non-transitory, computer-readable medium of claim 13 , wherein the instructions, when executed by the computer, cause the computer to:
classify each of the plurality of co-occurrence matrices into one of a plurality of categories, wherein individual values of the plurality of weight values are each associated with one of a plurality of categories; and
apply the one of the plurality of weight values to each of the plurality of co-occurrence matrices by applying the individual value of the plurality of weight values corresponding to the one of the plurality of categories into which the co-occurrence matrix was classified.