IMPROVED DOUBLE-LEVEL BLENDING MODEL (IDLBM): CUSTOMER CHURN ESTIMATING IN INSURANCE INDUSTRY

The simplest definition of churn is when a customer quits making purchases from a company. Churn is frequently measured by the rate or share of customers who drop a brand over time. To increase the reliability of client churn forecast in the insurance market and improve the forecast of client attrition in the insurance industry, an Improved Double-Level Blending model (IDLBM) based on Dynamic integration is developed. The Random Forest (RF), Multi-layer Perceptron(MLP), Support Vector Machine(SVM), and Bayesian Network (BN) are used to find the Insurance industry customer churn forecast. all in accordance with the features of client dataset from the insurance business. The dataset insurance Churn Prediction-Machine Hackthan was taken from the Kaggle's repository. The dataset is evaluated using two separate strategies. First, there is traditional technique Blending, second, there is IDLBM, which multiplies the outcomes of each classifier's predictions to improve the learner's prediction accuracy performance. The investigational outcomes show that IDLBM may greatly increase the accuracy of Insurance sector consumer loss prediction.