# US10984913B2: Blockchain System For Natural Language Processing

- Kind: Granted patent
- Family: Blockchain System For Natural Language Processing (https://www.intellectualfrontiers.com/patents/blockchain-system-for-natural-language-processing)
- USPTO status: Patented Case (as of 2021-03-31; record read 2026-09-27)
- Application: 15389243
- Filed: 2016-12-22
- Published: 2017-04-13
- Granted: 2021-04-20
- Estimated expiration: 2034-09-05 (An estimate from Google Patents, read 9 October 2026. Fees, terminal disclaimers and later events can change it, and it is not legal advice.)
- Inventor: Shahid N. Shah
- Rights holder: Intellectual Frontiers LLC
- Classifications (CPC): G16H 50/20, G16H 10/60, G06F 16/951, G06F 40/40, G06Q 10/40, G06F 11/1464, G06F2201/80, G06F2201/84, G06Q 10/101
- Continuation-in-part of US20130290022A1

## Licence or acquisition

Available for an exclusive or non-exclusive licence, field-of-use licences included, or for acquisition. Terms on inquiry. Write to hello+ip@intellectualfrontiers.com.

## In plain language

[Blockchain-Powered Healthcare Data for Natural Language Processing (NLP)- Collection, Storage, and Standardization](https://www.intellectualfrontiers.com/patent-summaries/blockchain-powered-healthcare-data-for-natural-language-processing-nlp-collection-storage-and-standardization)

## Abstract

A blockchain configured system includes a router and a blockchain configured record bank. The router collects data and converts it in a format in accordance with a defined standard. The blockchain configured record bank can include or be coupled to a data repository. The blockchain configured record bank can be configured to be coupled to the data provider through the router over a communication network. The blockchain configured record bank stores the data received from the data provider and can be accessible or searchable from within or outside the blockchain configured record bank. The blockchain configured record bank can be coupled to or include a data logging unit that maintains metadata associated with the data and configured to facilitate natural language processing capabilities. The router and the blockchain configured record bank may be coupled to machine learning system, metadata validation system, and master data validation system.

## Independent claims

1. A blockchain configured geographically distributed architecture-based system connected over a communication network for transforming an unstructured or semi-structured dataset to a structured computerized dataset fora blockchain configured records database communicatively coupled to a plurality of blockchain configured content based routers receiving said unstructured or semi-structured dataset from a plurality of data provider computers in a blockchain-enabled network, said system comprising: a first proxy database, stored on a first tangible non-transitory computer readable medium and comprising a first special purpose processing device implemented on a first integrated circuit chip, configured to: create a backup of data associated with a first data provider computer, wherein said data associated with said first data provider computer is in a first digital format; and communicate with said first data provider computer for backing up said data associated with said first data provider computer, through a first proxy object, wherein said first proxy object comprises one or more references to said first proxy database to establish a connection between said first data provider computer and said first proxy database through one or more database drivers; a blockchain configured first content based router of said plurality of blockchain configured content based routers, comprising a second special purpose processing device implemented on a second integrated circuit chip, configured to: collect said data associated with said first data provider computer from said first proxy database; and convert said data associated with said first data provider computer to said structured computerized dataset in accordance with a standardized digital format of datasets associated with said blockchain configured records database, wherein said blockchain configured first content based router is physically located at a gateway associated with said first data provider computer providing a first digital data access point of said distributed blockchain configured records database to said first data provider computer; said blockchain configured records database stored on a third tangible non-transitory computer readable medium and comprising a third special purpose processing device implemented on a third integrated circuit chip to store and index said structured computerized dataset in said standardized digital format of datasets associated with said blockchain configured records database, and providing a plurality of distributed digital data access points including said first digital data access point communicatively connecting with said first data provider computer through said first digital data access point of said plurality of distributed digital data access points; a master data validation system communicatively and operatively coupled to said blockchain configured first content based router and said blockchain configured records database and including a master data repository to store master data instances in said standardized digital format of datasets associated with said blockchain configured records database on a tangible non-transitory memory device; and a metadata validation system that is different and separately located from said master data validation system and communicatively and operatively coupled to said blockchain configured first content based router, and said master data validation system, and configured to store linkable metadata layer objects in a metadata layer repository and digitally link said metadata layer objects by storing an identifier to a source content residing in said blockchain configured records database, wherein said blockchain configured first content based router is connected to a machine learning system contained within said blockchain configured records database, said machine learning system comprising: internal extensible taxonomies built in a computerized format based on said structured computerized dataset in said blockchain configured records database, said internal extensible taxonomies are defined through a computerized category profile wherein said computerized category profile includes digitally stored parent terms and digitally stored child terms and associated digital identifiers indicative of said respective digitally stored parent terms, digitally stored child terms and pointers indicating mutual connections in a hierarchical pattern, wherein said internal extensible taxonomies are automatically generated through machine learning over a period of time, and vs/herein said machine learning system receives inputs from said master data validation system and said metadata validation system to build said internal extensible taxonomies; external taxonomies pulled from external systems not connected with said blockchain configured records database directly and crawled by accessing said external systems with the use of a search engine-enabled crawler to merge with said internal extensible taxonomies digitally by mapping comparable terms in said external taxonomies with said category profile of said internal extensible taxonomies; a memory circuit to store said internal extensible taxonomies and said external taxonomies that are digitally merged with said internal extensible taxonomies; and a semantics learning appliance comprising a fourth special purpose processing device implemented on a fourth integrated circuit chip configured to perform mapping of said inflowing unstructured or semi-structured dataset with said structured computerized dataset already stored in said blockchain configured records database and said master data and said metadata; wherein a dedicated first local machine housed in a material frame is operatively and communicatively coupled to said first data provider computer and installed physically behind a firewall and communicatively connected with said remotely located blockchain configured records database to transmit said structured computerized dataset at least in part from said blockchain configured records database to said first data provider computer through a digital transmission channel based on access rights and upon identity verification of said first local machine by said blockchain configured records database, and wherein said first local machine comprises: a first agent device configured to gather browser, device, and network information associated with said data provider computer and transmit said gathered browser, device, and network information to said blockchain configured records database for identity verification to authorize accessing of at least a portion of said structured computerized dataset stored in said blockchain configured records database, and a first contextual sensor fitted proximate to said first data provider computer physically for gathering contextual information about said first data provider computer and transmitting said contextual information to said blockchain configured records database for further identity verification to authorize accessing of said at least a portion of said structured dataset stored in said blockchain configured records database, wherein said first contextual sensor comprises a Global Positioning System-based (GPS-based) device configured to sense geo-spatial information indicative of location coordinates and transmit said sensed geo-spatial information to said blockchain configured records database for verification of identity, of said first data provider computer.

Source: https://www.intellectualfrontiers.com/patents/blockchain-system-for-natural-language-processing/us10984913b2