Cognitive systematic review (CSR) for smarter cognitive solutions

Patent metadata

Field	Value
Publication number	US-11526788-B2
Application number	US-201816004597-A
Country	US
Kind code	B2
Filing date	Jun 11, 2018
Priority date	Jun 11, 2018
Publication date	Dec 13, 2022
Grant date	Dec 13, 2022

Abstract

An approach for determining a veracity of a reported event is provided. In an embodiment, a set of predictor variables is retrieved from a selected use case. Each of these predictor values is a condition that indicates the veracity of the reported event. In addition, a set of hidden predictor variables is generated from a set of unstructured documents related to the reported event using a hidden Markov model that is based on the predictor variables using a cognitive system. These hidden predictor variables are combined with the set of predictor variables to generate a set of updated predictor variables. These updated predictor variables are used by the cognitive system to return a determination of the veracity of the reported event.

First claim

What is claimed is:

1. A method for determining a veracity of a reported event, comprising:
retrieving, from a selected use case selected from a plurality of use cases, a set of predictor variables specific to the selected use case, each predicator variable of the set of predictor variables being a condition that indicates the veracity of the reported event, wherein the reported event is claimed by a reporter to have previously occurred;
generating, using a cognitive system, a set of hidden predictor variables from a set of unstructured documents related to the reported event using a hidden Markov model based on a linear regression that is run on an extension of the predictor variables;
generating a set of updated predictor variables by a combining of the set of predictor variables and the hidden predictor variables that includes any newly discovered dependencies between predictor variables; and
returning a determination of the veracity of the reported event based on an analysis by the cognitive system using the updated predictor variables of event details of the reported event provided by the reporter along with an identification of the updated predictor variables used to make the determination.

2. The method of claim 1 , the method further comprising:
comparing, using a cognitive system, the set of event details with a plurality of use cases in a use case database;
selecting a selected use case based on the comparing.

3. The method of claim 2 , the method further comprising adding a use case having the updated predictor values to the use case database.

4. The method of claim 2 , further comprising scoring the unstructured documents for relevance based on the set of event details.

5. The method of claim 4 , wherein the set of unstructured documents includes unstructured social media data, which includes any location information of a person associated with the reported event, statements of the person about the reported event, and environmental conditions occurring in a location of the reported event, and unstructured research article data that includes articles that focus on gathering detailed non-numerical data pertaining to a particular set of events and extrapolating from the non-numerical data to explain a particular phenomenon.

6. The method of claim 1 , further comprising assigning, by the cognitive system, a weight to each hidden predictor variable of the set of hidden predictor variables based on a predictive ability of the hidden predictor variable.

7. The method of claim 1 , wherein the reported event is an insurance claim and wherein the returning of the determination of veracity determines whether the insurance claim is fraudulent.

8. A system for determining a veracity of a reported event, comprising:
a memory medium comprising instructions;
a bus coupled to the memory medium; and
a processor coupled to the bus that when executing the instructions causes the system to:
retrieve, from a selected use case selected from a plurality of use cases, a set of predictor variables specific to the selected use case, each predicator variable of the set of predictor variables being a condition that indicates the veracity of the reported event, wherein the reported event is claimed by a reporter to have previously occurred;
generate, using a cognitive system, a set of hidden predictor variables from a set of unstructured documents related to the reported event using a hidden Markov model based on a linear regression that is run on an extension of the predictor variables;
generate a set of updated predictor variables by a combining of the set of predictor variables and the hidden predictor variables that includes any newly discovered dependencies between predictor variables; and
return a determination of the veracity of the reported event based on an analysis by the cognitive system using the updated predictor variables of event details of the reported event provided by the reporter with an identification of the updated predictor variables used to make the determination.

9. The system of claim 8 , the instructions further causing the system to:
compare, using the cognitive system, the set of event details with a plurality of use cases in a use case database;
select a selected use case based on the comparing.

10. The system of claim 9 , the instructions further causing the system to add a use case having the updated predictor values to the use case database.

11. The system of claim 9 , the instructions further causing the system to score the unstructured documents for relevance based on the set of event details.

12. The system of claim 11 , wherein the set of unstructured documents includes unstructured social media data, which includes any location information of a person associated with the reported event, statements of the person about the reported event, and environmental conditions occurring in a location of the reported event, and unstructured research article data that includes articles that focus on gathering detailed non-numerical data pertaining to a particular set of events and extrapolating from the non-numerical data to explain a particular phenomenon.

13. The system of claim 8 , the instructions further causing the system to assign, by the cognitive system, a weight to each hidden predictor variable of the set of hidden predictor variables based on a predictive ability of the hidden predictor variable.

14. The system of claim 8 , wherein the reported event is an insurance claim and wherein the returning of the determination of veracity determines whether the insurance claim is fraudulent.

15. A computer program product for determining a veracity of a reported event, the computer program product comprising a computer readable storage media, and program instructions stored on the computer readable storage media, that cause at least one computer device to:
retrieve, from a selected use case selected from a plurality of use cases, a set of predictor variables specific to the selected use case, each predicator variable of the set of predictor variables being a condition that indicates the veracity of the reported event, wherein the reported event is claimed by a reporter to have previously occurred;
generate, using a cognitive system, a set of hidden predictor variables from a set of unstructured documents related to the reported event using a hidden Markov model based on a linear regression that is run on an extension of the predictor variables;
generate a set of updated predictor variables by a combining of the set of predictor variables and the hidden predictor variables that includes any newly discovered dependencies between predictor variables; and
return a determination of the veracity of the reported event based on an analysis by the cognitive system using the updated predictor variables of event details of the reported event provided by the reporter along with an identification of the updated predictor variables used to make the determination.

16. The computer program product of claim 15 , the instructions further causing the at least one computer device to:
compare, using the cognitive system, the set of event details with a plurality of use cases in a use case database;
select a selected use case based on the comparing.

17. The computer program product of claim 15 , the instructions further causing the at least one computer device to add a use case having the updated predictor values to the use case database.

18. The computer program product of claim 16 , the instructions further causing the at least one computer device to

Assignees

• Kyndryl Inc

Inventors

• Zolotow Clea
• Lawrence Calvin D
• Northway Tedrick N
• Delaney John
• Iqbal Mickey

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06N5/041

  Abduction · CPC title

• G06N7/08

  using chaos models or non-linear system models · CPC title

• G06N20/00 Primary

  Machine learning · CPC title

• G06N5/048 Primary

  Fuzzy inferencing · CPC title