Multi-objective rain optimization algorithm with WELM model for customer churn prediction in telecommunication sector

Abstract Customer retention is a major challenge in several business sectors and diverse companies identify the customer churn prediction (CCP) as an important process for retaining customers. CCP telecommunication sector has become essential need owing to rise number of service providers. Recently, machine learning (ML) deep (DL) models have begun develop effective model. This paper presents new improved synthetic minority over-sampling technique (SMOTE) with optimal weighted extreme (OWELM) called ISMOTE-OWELM model CCP. The presented comprises preprocessing, balancing unbalanced dataset, classification. multi-objective rain optimization algorithm (MOROA) used two purposes: determining sampling rate SMOTE parameter tuning WELM. Initially, data involve normalization class labeling. Then, ISMOTE employed handle imbalanced dataset where (ROA) applied determine rate. At last, WELM labels data. Extensive experimentation carried out ensure against Telecommunication dataset. simulation outcome portrayed that superior other accuracy 0.94, 0.92, 0.909 on I, II, III, respectively.