# US 2020/0401716 A1 · Utilizing neural networks for data anonymization > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2020/0401716 A1 - Document ID: US20200401716A1 - Status: Published application - Kind: A1 - Application number: 16/902,937 - Filed: 2020-06-16 - Published: 2020-12-24 - Earliest claimed: 2019-06-18 - Assignees: Yva.Ai, Inc. - Applicants: Yva.ai, Inc. - Inventors: David Yan; Victor Kuznetsov; Anton Kirillov; Viacheslav Seledkin - Web page: https://patentgrep.com/patent/US20200401716A1 ## CPC - 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} - G06N 3/02: Neural networks - G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] - G06N 3/045: Combinations of networks - G06N 3/0455: Auto-encoder networks; Encoder-decoder networks - G06N 3/088: Non-supervised learning, e.g. competitive learning - G06N 3/09: Supervised learning - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract An example method comprises: receiving a natural language text; transforming, by a neural network, the natural language text into a numeric representation comprising a plurality of numeric values; discarding the natural language text; and performing, using the numeric representation of the natural language text, an information extraction task. ## Claims ### Claim 1 A method, comprising: receiving, by a computer system, a natural language text; producing, by a neural network, a numeric representation of the natural language text; discarding the natural language text; and performing, using the numeric representation of the natural language text, an information extraction task. ### Claim 2 (depends on 1) The method of claim 1, wherein the neural network is an encoder part of an autoencoder. ### Claim 3 (depends on 1) The method of claim 1, wherein producing the numeric representation of the natural language text further comprises: applying an irreversible distorting transformation to the numeric representation of the natural language text. ### Claim 4 (depends on 3) The method of claim 3, wherein applying to the numeric representation of the natural language text the irreversible distorting transformation further comprises: adding, to each element of the numeric representation, a sample of a specified statistical distribution. ### Claim 5 (depends on 1) The method of claim 1, wherein performing the information extraction task comprises: associating the natural language text with a category of a predefined set of categories. ### Claim 6 (depends on 1) The method of claim 1, wherein performing the information extraction task comprises: evaluating a process characterized by the natural language text. ### Claim 7 (depends on 1) The method of claim 1, wherein performing the information extraction task comprises: evaluating a likelihood of occurrence of a specified event or condition. ### Claim 8 (depends on 1) The method of claim 1, wherein performing the information extraction task comprises: evaluating an emotional aspect of the natural language text. ### Claim 9 (depends on 1) The method of claim 1, wherein performing the information extraction task comprises: feeding the numeric representation of the natural language text to a neural network that produces a numeric value indicative of a characteristic of the natural language text. ### Claim 10 A system, comprising: a memory; and a processor coupled to the memory, the processor configured to: receive a natural language text; transform, by a neural network, the natural language text into a numeric representation; discard the natural language text; and perform, using the numeric representation of the natural language text, an information extraction task. ### Claim 11 (depends on 1) The system of claim 1, wherein the neural network is an encoder part of an autoencoder. ### Claim 12 (depends on 1) The system of claim 1, further comprising: applying, to the numeric representation of the natural language text, an irreversible distorting transformation. ### Claim 13 (depends on 1) The system of claim 1, wherein performing the information extraction task comprises: associating the natural language text with a category of a predefined set of categories. ### Claim 14 (depends on 1) The system of claim 1, wherein performing the information extraction task comprises: evaluating a process characterized by the natural language text. ### Claim 15 (depends on 1) The system of claim 1, wherein performing the information extraction task comprises: evaluating a likelihood of occurrence of a specified event or condition. 16 The system of claim 1, wherein performing the information extraction task comprises: evaluating an emotional aspect of the natural language text. ### Claim 17 A non-transitory computer-readable storage medium comprising executable instructions that, when executed by a computer system, cause the computer system to: receive a natural language text; transform, by a neural network, the natural language text into a numeric representation; discard the natural language text; and perform, using the numeric representation of the natural language text, an information extraction task. ### Claim 18 (depends on 17) The non-transitory computer-readable storage medium of claim 17, wherein the neural network is an encoder part of an autoencoder. ### Claim 19 (depends on 17) The non-transitory computer-readable storage medium of claim 17, further comprising executable instructions that, when executed by the computer system, cause the computer system to: apply, to the numeric representation of the natural language text, an irreversible distorting transformation. ### Claim 20 (depends on 17) The non-transitory computer-readable storage medium of claim 17, wherein performing the information extraction task comprises: feeding the numeric representation of the natural language text to a neural network that produces a numeric value indicative of a characteristic of the natural language text.