Life Insurance Prediction and Its Sustainability Using Machine Learning Approach

Owning life insurance coverage that is not enough to pay for the expenses called underinsurance, and it has been found have a significant influence on sustainability financial health of families. However, companies need good profile potential policyholders. Customer profiling become one essential marketing strategies any sustainable business, such as market, identify purchasers. One well-known method carrying out customer segmenting machine learning. Hence, this study aims provide helpful framework predicting policyholders using data mining approach with different sampling methods lead transition industry development. Various samplings, Synthetic Minority Over-sampling Technique, Randomly Under-Sampling, ensemble (bagging boosting) techniques, are proposed handle imbalanced dataset. The result reveals decision tree best performer according ROC and, balanced accuracy, F1 score, GM comparison, Naïve Bayes seems be performer. It also models do guarantee high performance in ensembled plays role overcoming problem.