Granted Patent B2

US 12,489,611 · App. 18/142,120 · Granted Dec 2, 2025

Inventors: Radhika Vinaikrishnan Nair (Maharashtra, IN); Dalydas Karanath (Kerala, IN); Syed Najumudeen (Telangana, IN)

Assignee: BANK OF AMERICA CORPORATION

H04L9/0852G06N10/20

Abstract

Claims (91)

1 . A system for dynamic folding and compression of data using quantum computing, the system comprising:

a memory device with computer-readable program code stored thereon;

at least one processing device operatively coupled to the at least one memory device and the at least one communication device, wherein executing the computer-readable code is configured to cause the at least one processing device to perform the following operations:

determine a list of similarities through a comparison between a dataset and a historical record associated with previous datasets,

wherein determination of the list of similarities of the dataset comprises assessing the dataset for similarities to the historical record and mapping the dataset to a process map table;

recreate, in response to the list of similarities of the dataset, the dataset through the process map table;

compress the dataset based on similarities found in the historical record;

determine a folding type for folding of the compressed dataset;

fold the compressed dataset using the determined folding type,

wherein folding the compressed dataset comprises creating a dynamic hash table with a key value data pair associated with the folded and compressed dataset; and

encrypt the folded and compressed dataset, the dynamic hash table, and the key value data pair through quantum computing utilizing linear compression.

2 . The system of claim 1 , wherein the determination of the folding type for compression causes the at least one processing device to perform the following operations:

apply a first convolution on the compressed dataset to create a first set of output channels;

apply a max pooling function on the first set of output channels to create a first set of max pooled output channels;

apply a second convolution on the first set of max pooled output channels to create a second set of output channels;

apply a second max pooling function on the second set of output channels to create a second set of max pooled output channels;

apply a rectified linear unit (ReLU) activation function on the second set of max pooled output channels to create a folding compatibility output; and

apply a soft max function on the folding compatibility output to determine the folding type for the compressed dataset.

3 . The system of claim 2 , wherein the application of the max pooling function comprises reduction of dimensionality of an input through filtration of the input to collect a determined largest amount of data within a channel.

4 . The system of claim 1 , wherein determination of the list of similarities of the dataset further comprises generative pre-trained transformer framework to assess the historical record for corpora that can be compressed through the process map table.

5 . The system of claim 1 , wherein recreation of the dataset through the process map table further comprises clustering the dataset and performing a lexical analysis.

6 . The system of claim 1 , wherein encryption of the folded and compressed dataset and the key value data pair through quantum computing utilizing linear compression causes the at least one processing device to perform the following operations:

eliminate, via a structural redundant elimination, a post folded set of redundant data from the folded and compressed dataset, the dynamic hash table, and the key value data pair;

convert the folded and compressed dataset, the dynamic hash table, and the key value data pair into a quantum state representation; and

encrypt the quantum state representation using linear similarities found within the quantum state representation.

7 . The system of claim 6 , wherein the encrypted quantum state representation is transmitted to a predetermined destination.

8 . The system of claim 6 , wherein an inverse encryption function, an unfolding, and a decompression of the encrypted quantum state representation causes the at least one processing device to perform the following operations:

crack the encrypted quantum state representation encryption using the inverse encryption function on the encrypted quantum state representation;

obtain the folded and compressed dataset, the dynamic hash table, and the key value data pair from the cracked encrypted quantum state representation;

unfold the folded and compressed dataset using the dynamic hash table and the key value data pair;

obtain the compressed dataset from the unfolding of the folded and compressed dataset;

decompress the compressed dataset; and

obtain the dataset through the decompression of the compressed dataset.

9 . A computer program product for dynamic folding and compression of data using quantum computing, wherein the computer program product comprises at least one non-transitory computer-readable program code portion embodied therein, the computer-readable program code portion which when executed by a processing device is configured to cause the processor to perform the following operations:

determine a list of similarities through a comparison between a dataset and a historical record associated with previous datasets,

wherein determination of the list of similarities of the dataset comprises assessing the dataset for similarities to the historical record and mapping the dataset to a process map table;

recreate, in response to the list of similarities of the dataset, the dataset through the process map table;

compress the dataset based on similarities found in the historical record;

determine a folding type for folding of the compressed dataset;

fold the compressed dataset using the determined folding type,

wherein folding the compressed dataset comprises creating a dynamic hash table with a key value data pair associated with the folded and compressed dataset; and

encrypt the folded and compressed dataset, the dynamic hash table, and the key value data pair through quantum computing utilizing linear compression.

10 . The computer program product of claim 9 , wherein in the determination of the folding type for compression, the processing device is further configured to cause the processor to perform the following operations:

apply a first convolution on the compressed dataset to create a first set of output channels;

apply a max pooling function on the first set of output channels to create a first set of max pooled output channels;

apply a second convolution on the first set of max pooled output channel to create a second set of max pooled output channels;

apply a second max pooling function on the second set of output channels to create a second set of max pooled output channels;

apply a rectified linear unit (ReLU) activation function on the second set of max pooled output channels to create a folding compatibility output; and

apply a soft max function on the folding compatibility output to determine the folding type for the compressed dataset.

11 . The computer program product of claim 10 , wherein the application of the max pooling function comprises reduction of dimensionality of an input through filtration of the input to collect a determined largest amount of data within a channel.

12 . The computer program product of claim 9 , wherein determination of the list of similarities of the dataset further comprises generative pre-trained transformer framework to assess the historical record for corpora that can be compressed through the process map table.

13 . The computer program product of claim 9 , wherein recreation of the dataset through the process map table further comprises clustering the dataset and performing a lexical analysis.

14 . The computer program product of claim 9 , wherein encryption of the folded and compressed dataset and the encrypted key value data pair through quantum computing utilizing linear compression causes the processing device to further be configured to cause the processor to perform the following operations:

eliminate, via a structural redundant elimination, a post folded set of redundant data from the folded and compressed dataset, the dynamic hash table, and the key value data pair;

convert the folded and compressed dataset, the dynamic hash table, and the encrypted key value data pair into a quantum state representation; and

encrypt the quantum state representation using linear similarities found within the quantum state representation.

15 . The computer program product of claim 14 , wherein an inverse encryption function, an unfolding, and a decompression the encrypted quantum state representation causes the processing device to further be configured to cause the processor to perform the following operations:

crack the encrypted quantum state representation encryption using the inverse encryption function on the encrypted quantum state representation;

obtain the folded and compressed dataset, the dynamic hash table, and the key value data pair from the cracked encrypted quantum state representation;

unfold the folded and compressed dataset using the dynamic hash table and the key value data pair;

obtain the compressed dataset from the unfolding of the folded and compressed dataset;

decompress the compressed dataset; and

obtain the dataset through the decompression of the compressed dataset.

16 . A computer-implemented method for dynamic data folding and compression of data using quantum computing, the computer-implemented method comprising:

determining a list of similarities through a comparison between a dataset and a historical record associated with previous datasets;

wherein determining the list of similarities of the dataset comprises assessing the dataset for similarities to the historical record and mapping the dataset to a process map table;

recreating, in response to the list of similarities of the dataset, the dataset through the process map table;

compressing the dataset based on similarities found in the historical record;

determining a folding type for folding of the compressed dataset; and

folding the compressed dataset using the determined folding type,

wherein folding the compressed dataset comprises creating a dynamic hash table with a key value data pair associated with the folded and compressed dataset; and

encrypting the folded and compressed dataset, the dynamic hash table, and the key value data pair through quantum computing utilizing linear compression.

17 . The computer-implemented method of claim 16 , wherein the determination of the folding type for compression further comprises:

applying a first convolution on the compressed dataset to create a first set of output channels;

applying a max pooling function on the first set of output channels to create a first set of max pooled output channels;

applying a second convolution on the first set of max pooled output channels to create a second set of output channels;

applying a second max pooling function on the second set of output channels to create a second set of max pooled output channels;

applying a rectified linear unit (ReLU) activation function on the second set of max pooled output channels to create a folding compatibility output; and

applying a soft max function on the folding compatibility output to determine the folding type for the dataset.

18 . The method of claim 17 , wherein the application of the max pooling function comprises reduction of dimensionality of an input through filtration of the input to collect a determined largest amount of data within a channel and eliminate smaller data within the channel.

19 . The method of claim 16 , wherein encryption of the folded and compressed dataset, the dynamic hash table, and the key value data pair through quantum computing utilizing linear compression further comprises:

eliminating, via a structural redundant elimination, a post folded set of redundant data from the folded and compressed dataset, the dynamic hash table, and the key value data pair;

converting the folded and compressed dataset, the dynamic hash table, and the encrypted key value data pair into a quantum state representation; and

encrypting the quantum state representation using linear similarities found within the quantum state representation.

20 . The method of claim 19 , wherein an inverse encryption function, an unfolding, and a decompression of the encrypted quantum state representation comprises;

cracking the encrypted quantum state representation encryption using the inverse encryption function on the encrypted quantum state representation;

obtaining the folded and compressed dataset, dynamic hash table, and key value data pair from the cracked encrypted quantum state representation;

unfolding the folded and compressed dataset using the dynamic hash table, and the key value pair;

obtaining the compressed dataset from the unfolding of the folded and compressed dataset;

decompressing the compressed dataset; and

obtain the dataset through the decompression of the compressed dataset.

Assignments (1)

Continuity (1)

References Cited (13)