ABSTRACT
Trying to predict customer churn is most prevalent challenging tasks in just about any industry. With breakthroughs in ML and AI, the capability to forecast client attrition has grown significantly. This paper incorporates the majority of the algorithms developed for customer churn prediction techniques. We studied a number of papers published by renowned authors. We attempted to document of all the machine learning and deep learning methodologies invented and deployed by the world's technological behemoths to better understand their clientele and expand the market for their products or services. According to the findings, the DL model outperformed the competition in terms of classification and prediction accuracy. DL models would thus choose to disregard such as useless information while constructing their data blueprints. This review paper explains how to use a Deep Learning, ML and senti churn approach for predict churn on a company metadata. The churn prediction model takes into account contextual elements, usage characteristics, customer attributes, and support characteristics. Deep learning and machine learning modeling techniques have been trained using historical datasets.
