A Fuzzy Random Survival Forest for Predicting Lapses in Insurance Portfolios Containing Imprecise Data

We propose a fuzzy random survival forest (FRSF) to model lapse rates in life insurance portfolio containing imprecise or incomplete data such as missing, outlier, noisy values. Following the methodology, FRSF is proposed new machine learning technique for solving time-to-event using an ensemble of multiple trees. In process, combination methods c-index, sets theory, and trees enable automatic handling data. analyse results several experiments test them statistically; they show FRSF’s robustness, verifying that its generalisation capacity not reduced when modelling Furthermore, obtained real company demonstrate has better performance comparison with other state-of-the-art algorithms traditional Cox tree-based techniques forest.