Browse Prior Art Database

Watson Cognitive Disability Claims Risk Scoring

IP.com Disclosure Number: IPCOM000247534D
Publication Date: 2016-Sep-14
Document File: 4 page(s) / 241K

Publishing Venue

The IP.com Prior Art Database

Abstract

A WATSON based artificial intelligence enabled method and process for implementing Cognitive Disability Insurance Claims Data Profiler and Risk Score generator using smart aggregation and integration of both unstructured data from disparate data sources and structured data from industry standard regulatory data sources is disclosed. This disclosure defines the implementable method and steps for providing the below listed Disability Insurance Claims processing capabilities: 1) Cognitive Evaluation / Profiling and Enrichment of Aggregated Social Media Data 2) Cognitive Evaluation / Profiling and Enrichment of Criminal Records for matching Disability Claimant persona 3) Cognitive Evaluation of Financial Risks (Credit Score) and involvement in other Financial Scams 4) Cognitive tag generation which can be converted to a number based on AI platform capabilities (e.g. Number of Data Associations, Artificial Intelligence based generated questions, Predictions etc) 5) Cognitive enabled Disability Claims evaluation algorithm with expert rules and scoring mechanism based on proprietary business weights.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 48% of the total text.

Page 01 of 4

Watson Cognitive Disability Claims Risk Scoring

Currently, disability insurance carriers around the world are following an expert judgement based claims handling process. This process has the following limitations:
1) Claim Leakages - Fraud claims are processed and paid out resulting in heavy payouts
2) Non-Efficient Claimant Risk Profiling - No available solution supporting identification of risk patterns based mapping and scoring for the claimant using data available in multiple systems like Police Database, Credit Scores and Social Media Data etc.

This disclosure will help Disability Insurance carriers to cognitively analyze the disability insurance claims. This solution also generates a risk score based on multiple data dimensions of a disability claimant and enables disability insurance carriers to take a decision.

This disclosure defines a prescriptive cognitive enabled risk scoring mechanism to analyze the risk profile of a disability claimant and provides the following benefits for the disability insurance carriers:
1) Provides a quantifiable measure for fast and accurate disability claims handling
2) Decision support tool for automated claims triage after claims registration.

Conceptual View of the Solution

Technical Solution Component Layers
1) Data Capture Layer - User is allowed to enter the following disability claim details in a web page (e.g. Customer Portal etc)

Name, Age, Sex, Disability Reason / Cause, Driver's License #, Employment Information, House Address, Employer Address, Supervisor Name, Supervisor Designation, Injury Time, Pre-existing Illness / Medical Information and other details. The user can also upload the supporting documents (e.g. Medical Reports,

1


Page 02 of 4

Disability Certifications, MRI Reports, Disability Photographs etc)


2) Watson Cognitive Layer - This layer will enable the following activities. The below diagram describes the multiple conceptual Watson cognitive solution layers that will be implemented in the disclosure.


a) Aggregate Claimant related data from the following data sources (both structured and unstructured data sources)

- OFAC Data Source - Provides police verification data including any prior felonies, DUI's, Criminal cases, Citations etc
- Credit Score Data Source - Provides the credit score of the Claimant
- Social Media Data Source - Provides the aggregated social media profile for the claimant with the following attribute values


- Risky Behavior
- Demographic Risks
- Social Behavior Risk
- Criminal Behavior Risk
- Moral Behavior Risk


b) Watson cognitive explorer will execute a cognitive pattern matching (e.g. Text Analytics, Buzz words  tracking etc) of already available criminals data, financial fraud data and social media fraud data of  known criminals and cases against the disability claimant profile to arrive at a cognitive tag (e.g. Name  Value Pair of Claimant name with the cognitive tag value).
c) Watson solution ...