Machine learning model modification and natural language processing

Patent metadata

Field	Value
Publication number	US-10846485-B2
Application number	US-201916577793-A
Country	US
Kind code	B2
Filing date	Sep 20, 2019
Priority date	Jan 10, 2018
Publication date	Nov 24, 2020
Grant date	Nov 24, 2020

Abstract

A system, computer program product, and method are provided to automate a framework for knowledge graph based persistence of data, and to resolve temporal changes and uncertainties in the knowledge graph. Natural language understanding, together with one or more machine learning models (MLMs), is used to extract data from unstructured information, including entities and entity relationships. The extracted data is populated into a knowledge graph. As the KG is subject to change, the KG is used to create new and retrain existing machine learning models (MLMs). Weighting is applied to the populated data in the form of veracity value. Blockchain technology is applied to the populated data to ensure reliability of the data and to provide auditability to assess changes to the data.

First claim

What is claimed is:

1. A computer system comprising:
an artificial intelligence platform, in communication with a processing unit;
a knowledge engine operatively coupled to the processing unit to train a machine learning model (MLM), the knowledge engine configured to:
receive natural language (NL) input and query the input against a first knowledge graph (KG), and extract one or more triplets from the first KG;
apply a selected MLM to a second KG different from the first KG, and extract one or more triplets from the second KG, wherein each triplet includes a subject, object, and a relationship;
for each extracted triplet:
obtain a blockchain (BC) identifier associated with each triplet; and
identify a triplet veracity value from a corresponding BC ledger;

detect a modification of the first KG from the extracted one or more triplets from the second KG;
evaluate the detected modification, including employ the obtained BC identifier to assess veracity of the detected modification; and
establish a link between the first KG and the second KG, wherein the link creates a relationship between the first KG and the second KG.

2. The system of claim 1 , further comprising the knowledge engine to:
identify one or more entities and relationships present in the first KG and absent from the selected MLM; and
create the identified entities and relationships as consumable data.

3. The system of claim 2 , further comprising the knowledge engine to stream the consumable data to update the structure of the selected MLM, and store the updated selected MLM as a new MLM.

4. The system of claim 1 , further comprising the knowledge engine to compare the linked first KG and second KG, the comparison comprising an evaluation of corresponding one or more veracity value components.

5. The system of claim 4 , wherein a conflict between two or more data elements in the linked first KG and second KG are identified and selectively replaced based on at least one of the one or more veracity value components.

6. A computer program product to process natural language (NL), the computer program product comprising a computer readable storage device having program code embodied therewith, the program code executable by a processing unit to:
receive NL input and query the input against a first knowledge graph (KG), and extract one or more triplets from the first KG;
apply a selected machine learning model (MLM) to a second KG different from the first KG, and extract one or more triplets from the second KG, wherein each triplet includes a subject, object, and a relationship, and for each extracted triplet:
obtain a blockchain (BC) identifier associated with each triplet; and
identify a triplet veracity value from a corresponding BC ledger;

detect a modification of the first KG from the extracted one or more triplets from the second KG;
evaluate the detected modification, including employ the obtained BC identifier to assess veracity of the detected modification; and
establish a link between the first KG and the second KG, wherein the link creates a relationship between the first KG and the second KG.

7. The computer program product of claim 6 , further comprising the program code to:
identify one or more entities and relationships present in the first KG and absent from the selected MLM; and
create the identified entities and relationships as consumable data.

8. The computer program product of claim 7 , further comprising the program code to stream the consumable data to update the structure of the selected MLM, and store the updated selected MLM as a new MLM.

9. The computer program product of claim 6 , further comprising the program code to compare the linked first KG and second KG, the comparison comprising an evaluation of corresponding one or more veracity value components.

10. The computer program product of claim 9 , wherein a conflict between two or more data elements in the linked first KG and second KG are identified and selectively replaced based on at least one of the one or more veracity value components.

11. A method for processing natural language (NL), comprising:
receiving NL input and querying the input against a first knowledge graph (KG), and extracting one or more triplets from the first KG;
applying a selected machine learning model (MLM) to a second KG different from the first KG, and extracting one or more triplets from the second KG, wherein each triplet includes a subject, object, and a relationship, and for each extracted triplet:
obtaining a blockchain (BC) identifier associated with each triplet; and
identifying a triplet veracity value from a corresponding BC ledger;

detecting a modification of the first KG from the extracted one or more triplets from the second KG;
evaluating the detected modification, including employing the obtained BC identifier to assess veracity of the detected modification; and
establishing a link between the first KG and the second KG, wherein the link creates a relationship between the first KG and the second KG.

12. The method of claim 11 , further comprising:
identifying one or more entities and relationships present in the first KG and absent from the selected MLM; and
creating the identified entities and relationships as consumable data.

13. The method of claim 12 , further comprising streaming the consumable data to update the structure of the selected MLM, and storing the updated selected MLM as a new MLM.

14. The method of claim 11 , further comprising comparing the linked first KG and second KG, the comparison comprising an evaluation of corresponding one or more veracity value components.

15. The method of claim 14 , wherein a conflict between two or more data elements in the linked first KG and second KG is identified and selectively replaced based on at least one of the one or more veracity value components.

16. The system of claim 1 , wherein the link maintains a structure of the first KG and the second KG.

17. The computer program product of claim 6 , wherein the link maintains a structure of the first KG and the second KG.

18. The method of claim 11 , wherein the link maintains a structure of the first KG and the second KG.

Assignees

• IBM

Inventors

• Bacarella David
• Barnebee Iv James H
• Lawrence Nicholas
• Patel Sumit

Classifications

• H04L9/50

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

• G06F40/20 Primary

  Natural language analysis (semantic analysis of natural language G06F40/30) · CPC title

• G06F40/40 Primary

  Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) · CPC title

• G06F40/295

  Named entity recognition · CPC title

• H04L9/3239

  involving non-keyed hash functions, e.g. modification detection codes [MDCs], MD5, SHA or RIPEMD · CPC title