TECHNICAL FIELD
The present invention relates to investigating insurance claims, in particular to
5 investigating insurance claims using machine learning based technique for
identifying claims that are high-risk and need to be investigated in more detail.
BACKGROUND
[0001] An insurance claim is a formal request to an insurance provider for
10 reimbursement against losses covered under an insurance policy. In life insurance,
effective claims management plays an important role for mitigating financial, legal,
and reputational risks associated with insurance provider. An effective claim
management also helps the insurance provider to assist the customer (or user or
insured) in a more effective way and to provide a claims experience that is digital,
15 simple and fast resolution.
WE CLAIM
1. A method for investigating insurance claims, the method comprising:
5 receiving a first set of parameters and a second set of parameters, wherein the
first set and the second set of parameters corresponds to a user;
inputting the first set and second set of parameters into a logistic regression
model;
determining a first probability of the insurance claims to decline based on an
10 output from the logistic regression model;
comparing the percentage of decline of insurance claims with a predetermined
threshold;
assigning an affinity score based on the result of comparison;
determining a second probability of the insurance claims to decline based on
15 the assigned affinity score; and
classifying the insurance claims into Red, amber, yellow and green (RAYG)
based on the determined first and the second probability.
2. The method as claimed in claim 1, wherein the first set of parameters are received
20 from a first source.
3. The method as claimed in claim 1, wherein the second set of parameters are
received from a second source.
25 4. The method as claimed in claim 1, wherein if the percentage of decline is more
than the predetermined threshold, the ass