Review on application of data mining in life insurance

Abstract

The growth of life insurance has been mainly depending on the risk of insured people. These risks are unevenly distributed among the people which can be captured from different characteristics and lifestyle. These unknown distribution needs to be analyzed from historical data and use for underwriting and policymaking in life insurance industry. Traditionally risk is calculated from selected features known as risk factors but today it becomes important to know these risk factors in high dimensional feature space. Clustering in high dimensional feature is a challenging task mainly because of the curse of dimensionality and noisy features. Hence the use of data mining and machine learning techniques should experiment to see some interesting pattern and behaviour. This will help life insurance company to protect from financial loss to the insured person and company as well. This paper focuses on analyzing hidden correlation among features and use it for risk calculation of an individual customer.