Figure 3. The comparison of different ML based models
Insurance-demographic risk can arise from many sources, each requires
accurate prediction modelling. For example, before providing a car insurance
product to an individual, the insurer needs to accurately predict the number
of claims the driver might make in the future.
Additionally, life-insurers need
to have an accurate estimate of the expected lifespan of demography before
providing life-insurance products to the individuals belonging to the
demography.
Claims modelling refers to the prediction of all future costs
associated with insurance claims made by the policyholders. Bi-variate
regression models can be used to model the dependency between the
frequency and severity of claims
.
Operational risk is the risk of direct or indirect loss resulting from
inadequate or failed interval processes, people, and systems or from external
events. Fraudulent Activities is one of the major sources of operational risk
for companies, particularly those in the finance sector. It can take on many
forms: for example, bank-activities related fraud (fraudulent credit card
transactions money laundering activities), and insurance-related fraud
(fraudulent insurance claims), securities and commodities fraud or other
frauds such as mass-marketing fraud or corporate fraud. Many studies in the
literature address fraud detection as a binary classification problem. Binary
classifiers such as logistic regression, neural networks, k-nearest neighbor,
decision trees and support vector machines are widely used in this section
of the literature.
THE 3 rd INTERNATIONAL SCIENTIFIC CONFERENCES OF STUDENTS AND YOUNG RESEARCHERS dedicated to the 99
th
anniversary of the National Leader of Azerbaijan Heydar Aliyev
126
