Immutable records based generation of machine learning models for dynamically triggering actions

Patent metadata

Field	Value
Publication number	US-11405459-B2
Application number	US-201916542339-A
Country	US
Kind code	B2
Filing date	Aug 16, 2019
Priority date	Mar 8, 2019
Publication date	Aug 2, 2022
Grant date	Aug 2, 2022

Abstract

Machine Learning (ML) models are deployed in digital platforms for data analytics. However, it is realized that there is growing trends of recognition that machine learning models expose new vulnerabilities in software systems, for instance training data poisoning, adversarial responses, model extraction, and the like. Embodiments of the present disclosure provide systems and methods for safeguarding training dataset by exploiting immutability feature and generating immutable machine learning models for data analytics. More specifically, immutable records of events are governed by smart contracts within highly secure permissioned distributed ledger. This dataset is used for training multiple machine learning models which are immutable in nature and further utilized for triggering actions for incoming request(s) from IoT platforms.

First claim

What is claimed is:

1. A processor implemented method, comprising:
receiving, via one or more hardware processors, an input source data corresponding to at least one trusted data source amongst a plurality of trusted data sources;
storing the input source data in a multiple validator node distributed ledger to obtain a plurality of immutable records, wherein the multiple validator node distributed ledger is created for storing the input source data based on at least one of a consensus algorithm and a smart contract between the plurality of trusted data sources;
generating one or more machine learning models using the plurality of immutable records retrieved from the multiple validator node distributed ledger;
storing the one or more machine learning models as one or more immutable machine learning models in the multiple validator node distributed ledger, wherein the one or more immutable machine learning models are generated based on the plurality of immutable records for one or more passive supervised machine learning techniques; and
dynamically triggering, one or more actions specific to at least one Internet of Things (IoT) device deployed in an IoT platform using the one or more immutable machine learning models based on an input request received corresponding to the at least one Internet of Things (IoT) device.

2. The processor implemented method of claim 1 , wherein the input request is received based on a triggered induced by at least one of a rule being executed or an event occurred in the IoT platform.

3. The processor implemented method of claim 1 , wherein the one or more actions comprise activation or deactivation of one or more functionalities of the at least one IoT device based on at least one of a rule being executed or an event occurred in the IoT platform.

4. The processor implemented method of claim 1 , further comprising:
receiving a request from one or more trusted data sources;
anonymizing the plurality of immutable records to obtain a set of anonymization data; and
dynamically retrieving from the multiple validator node distributed ledger and providing, at least one of (i) at least a subset of the set of anonymization data, and (ii) the one or more immutable machine learning models to the one or more trusted data sources, based on the request.

5. A system comprising:
a memory storing instructions;
one or more communication interfaces; and
one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware processors are configured by the instructions to:
receive an input source data corresponding to at least one trusted data source amongst a plurality of trusted data sources;
store the input source data in a multiple validator node distributed ledger to obtain a plurality of immutable records, wherein the multiple validator node distributed ledger is created for storing the input source data based on at least one of a consensus algorithm and a smart contract between the plurality of trusted data sources;
generate one or more machine learning models using the plurality of immutable records retrieved from the multiple validator node distributed ledger;
store the one or more machine learning models as the one or more immutable machine learning models in the multiple validator node distributed ledger, wherein the one or more immutable machine learning models are generated based on the plurality of immutable records for one or more passive supervised machine learning techniques; and
dynamically trigger, one or more actions specific to at least one Internet of Things (IoT) device deployed in an IoT platform using the one or more immutable machine learning models based on an input request received corresponding to the at least one Internet of Things (IoT) device.

6. The system of claim 5 , wherein the input request is received based on a triggered induced by at least one of a rule being executed or an event occurred in the IoT platform.

7. The system of claim 5 , wherein the one or more actions comprise activation or deactivation of one or more functionalities of the at least one IoT device based on at least one of a rule being executed or an event occurred in the IoT platform.

8. The system of claim 5 , wherein the one or more hardware processors are further configured by the instructions to:
receive a request from one or more trusted data sources;
anonymize the plurality of immutable records to obtain a set of anonymization data; and
dynamically retrieve from the multiple validator node distributed ledger and providing, at least one of (i) at least a subset of the set of anonymization data, and (ii) the one or more immutable machine learning models to the one or more trusted data sources, based on the request.

9. One or more non-transitory machine readable information storage media comprising one or more instructions that, when executed by one or more hardware processors, cause the one or more hardware processors to generate one or more immutable machine learning models for data analytics by:
receiving, via the one or more hardware processors, an input source data corresponding to at least one trusted data source amongst a plurality of trusted data sources;
storing the input source data in a multiple validator node distributed ledger to obtain a plurality of immutable records, wherein the multiple validator node distributed ledger is created for storing the input source data based on at least one of a consensus algorithm and a smart contract between the plurality of trusted data sources;
generating one or more machine learning models using the plurality of immutable records retrieved from the multiple validator node distributed ledger;
storing the one or more machine learning models as the one or more immutable machine learning models in the multiple validator node distributed ledger, wherein the one or more immutable machine learning models are generated based on the plurality of immutable records for one or more passive supervised machine learning techniques; and
dynamically triggering, one or more actions specific to at least one Internet of Things (IoT) device deployed in an IoT platform using the one or more immutable machine learning models based on an input request received corresponding to the at least one Internet of Things (IoT) device.

10. The one or more non-transitory machine readable information storage media of claim 9 , wherein the input request is received based on a triggered induced by at least one of a rule being executed or an event occurred in the IoT platform.

11. The one or more non-transitory machine readable information storage media of claim 9 , wherein the one or more actions comprise activation or deactivation of one or more functionalities of the at least one IoT device based on at least one of a rule being executed or an event occurred in the IoT platform.

12. The one or more non-transitory machine readable information storage media of claim 9 , wherein the instructions, when executed by the one or more hardware processors, further cause the one or more hardware processors to perform:
receiving a request from one or more trusted data sources;
anonymizing the plurality of immutable records to obtain a set of anonymization data; and
dynamically retrieving from the multiple validator node distributed ledger and providing, at least one of (i) at least a subset of the set of anonymization data, and (ii) the one or more immutable machine learning models to the one or more trusted data sources, based on the request.

Assignees

• Tata Consultancy Services Ltd

Inventors

• Panda Satanik
• Roy Choudhury Abhishek
• Misra Prateep
• Chattopadhyay Tanushyam

Classifications

• H04L9/50

  using hash chains, e.g. blockchains or hash trees · CPC title

• H04L63/12 Primary

  Applying verification of the received information (cryptographic mechanisms or cryptographic arrangements for data integrity or data verification H04L9/32) · CPC title

• H04L9/0637

  Modes of operation, e.g. cipher block chaining [CBC], electronic codebook [ECB] or Galois/counter mode [GCM] · CPC title

• G06N20/00

  Machine learning · CPC title

• G06F21/6254

  by anonymising data, e.g. decorrelating personal data from the owner's identification · CPC title