Open AccessComparison of Machine Learning Algorithms for Processing of Health Insurance Claim
Author(s) -
Paridhi Saxena,
Aakash Kumar Seth,
Gangesh Chawla,
Ranganath M. Singari
Publication year - 2020
Publication title -
international journal of advanced production and industrial engineering
Language(s) - English
Resource type - Journals
ISSN - 2455-8419
DOI - 10.35121/ijapie202004241
Subject(s) - profitability index , decision tree , machine learning , random forest , naive bayes classifier , product (mathematics) , algorithm , computer science , artificial intelligence , financial services , customer satisfaction , sample (material) , marketing , business , actuarial science , support vector machine , finance , mathematics , chemistry , chromatography , geometry
The health insurance industry protects against financial losses resulting from various health conditions. Since a long, it has relied on statistics and data to calculate risks and thereby, centre attention more profoundly on a particular target audience for increasing the operational efficiency of the industry. Technologies like Machine Learning and Artificial Intelligence prove to be an efficient tool for enabling insurance companies to predict the Customer Lifetime Value (CLV). This can be done using customer lifestyle behaviour data allowing to assess the customer's potential profitability for insurance companies. This creates a more personalised marketing offer within the audience. The insurance industry and its components constitute a dynamic and competitive sector representing approximately 2.7 percent of the US Gross Domestic Product (GDP). As customers have become progressively scrupulous about narrowing down their specific requirements, insurers and insurance companies are scrutinizing techniques for improving business operations and consumer satisfaction. An attempt in this regard has been made to analyse the “sample insurance claim prediction dataset" using various machine learning models including Decision tree, Random Forest algorithms, Naïve Bayes, K-nearest neighbour algorithm, Supper Vector machines and Neural Networks. A comparative analysis is performed to generate reports.