# US 12,182,179 B1 · Apparatus and methods for generating obfuscated data within a computing environment > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,182,179 B1 - Document ID: US12182179B1 - Status: Granted - Kind: B1 - Application number: 18/629,594 - Filed: 2024-04-08 - Granted: 2024-12-31 - Earliest claimed: 2024-04-08 - Assignees: nference, Inc. - Applicants: nference, Inc. - Inventors: Murali Aravamudan; Ajit Rajasekharan - Web page: https://patentgrep.com/patent/US12182179B1 ## CPC - G06F 16/285: {Clustering or classification} - G06F 21/6218: {to a system of files or objects, e.g. local or distributed file system or database} - G06F 21/6254: {by anonymising data, e.g. decorrelating personal data from the owner's identification} - G06F 40/20: Natural language analysis (semantic analysis of natural language G06F40/30) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract An apparatus for generating obfuscated data within a computing environment, comprising a processor and a memory containing instructions configuring the processor to access a database containing a plurality of private data elements belonging to at least a private record, generate a set of obfuscated data elements, representative of the at least a private record, as a function of the plurality of private data elements using an generative model, determine a first distance measure between at least an obfuscated data element within the set of obfuscated data elements and at least a private data element of the plurality of private data elements within the database, and verify the first distance measure is within a distance range, wherein a minimum threshold of the distance range is determined as a function of a deidentification parameter and a maximum threshold of the distance range is determined as a function of an obfuscation parameter. ## Claims ### Claim 1 An apparatus for generating obfuscated data within a computing environment, the apparatus comprising: at least a processor; and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to: access a database containing a plurality of private data elements belonging to at least a private record; generate, using a generative model a set of obfuscated data elements, representative of the at least a private record, as a function of the plurality of private data elements; determine a first distance measure between at least an obfuscated data element within the set of obfuscated data elements and at least a private data element of the plurality of private data elements within the database; and verify, for the at least an obfuscated data element within the set of obfuscated data elements, the first distance measure is within a distance range, wherein: a minimum threshold of the distance range is determined as a function of a deidentification parameter; and a maximum threshold of the distance range is determined as a function of an obfuscation parameter; cluster the plurality of private data elements within the database into a plurality of clusters using at least a clustering algorithm; select a subset of obfuscated data elements from the set of obfuscated data elements as a function of both the plurality of clusters and the verification; and transmit the subset of obfuscated data elements to one or more requesting entities within a computing environment. ### Claim 2 (depends on 1) The apparatus of claim 1, wherein generating the set of obfuscated data elements comprises sampling from a noise distribution on a deidentified version of the plurality of private data elements. ### Claim 3 (depends on 1) The apparatus of claim 1, wherein the generative model comprises a large language model (LLM). ### Claim 4 (depends on 1) The apparatus of claim 1, wherein the generative model comprises a transformer-based diffusion model. ### Claim 5 (depends on 1) The apparatus of claim 1, wherein the memory further contains instructions configuring the at least a processor to: fine-tune the generative model on a subset of the private data elements selected from the plurality of private data elements within the database corresponding to at least one pre-determined domain. ### Claim 6 (depends on 1) The apparatus of claim 1, wherein the deidentification parameter comprises a privacy protection level associated with at least one pre-determined domain. ### Claim 7 (depends on 1) The apparatus of claim 1, wherein the obfuscation parameter comprises an obfuscation risk tolerance level. ### Claim 8 (depends on 1) The apparatus of claim 1, wherein: at least a cluster of the plurality of clusters comprises a cluster centroid and a defined radius of influence; and selecting the subset of obfuscated data elements from the set of obfuscated data elements comprises: determining a second distance measure between the cluster centroid of at least a cluster of the plurality of clusters and at least a private data element of the plurality of private data elements within the database; verifying, for the cluster centroid of the at least a cluster of the plurality of clusters, the second distance measure is within the distance range; and selecting the subset of obfuscated data elements from the set of obfuscated data elements as a function of the verification. ### Claim 9 A method for generating obfuscated data within a computing environment, the method comprising: accessing, by at least a processor, a database containing a plurality of private data elements belonging to at least a private record; generating, by the at least a processor, a set of obfuscated data elements using a generative model as a function of the plurality of private data elements; determining, by the at least a processor, a first distance measure between at least an obfuscated data element within the set of obfuscated data elements and at least a private data element of the plurality of private data elements within the database; and verifying, for the at least an obfuscated data element within the set of obfuscated data elements, the first distance measure is within a distance range, wherein: a minimum threshold of the distance range is determined as a function of a deidentification parameter; and a maximum threshold of the distance range is determined as a function of an obfuscation parameter; cluster, by the at least a processor, the plurality of private data elements within the database into a plurality of clusters using at least a clustering algorithm; select, by the at least a processor, a subset of obfuscated data elements from the set of obfuscated data elements as a function of both the plurality of clusters and the verification; and transmit, by the at least a processor, the subset of obfuscated data elements to one or more requesting entities within a computing environment. ### Claim 10 (depends on 9) The method of claim 9, wherein generating the set of obfuscated data elements comprises sampling from a noise distribution on a deidentified version of the plurality of private data elements. ### Claim 11 (depends on 9) The method of claim 9, wherein the generative model comprises a large language model (LLM). ### Claim 12 (depends on 9) The method of claim 9, wherein the generative model comprises a transformer-based diffusion model. ### Claim 13 (depends on 9) The method of claim 9, further comprising: fine-tuning, by the at least a processor, the generative model on a subset of the private data elements selected from the plurality of private data elements within the database corresponding to at least one pre-determined domain. ### Claim 14 (depends on 9) The method of claim 9, wherein the deidentification parameter comprises a privacy protection level associated with at least one pre-determined domain. ### Claim 15 (depends on 9) The method of claim 9, wherein the obfuscation parameter comprises an obfuscation risk tolerance level. ### Claim 16 (depends on 9) The method of claim 9, wherein: at least a cluster of the plurality of clusters comprises a cluster centroid and a defined radius of influence; and selecting the subset of obfuscated data elements from the set of obfuscated data elements comprises: determining a second distance measure between the cluster centroid of at least a cluster of the plurality of clusters and the at least a private data element of the plurality of private data elements within the database; verifying, for the cluster centroid of the at least a cluster of the plurality of clusters, the second distance measure is within the distance range; and selecting the subset of obfuscated data elements from the set of obfuscated data.