Classification and Recommendations for Insurance Products Using Naïve Bayes Classification Based on Customer Data

The Digitalization Era was born with a series of benefits that were presented, such as being able to help companies to more easily reach their customers.Digitization is the increasing availability of digital data made possible by advances in creating, transferring, storing and analyzing digital data, and has the potential to structure, shape and influence the contemporary world.At the beginning of 2021 the insurance company had a huge impact due to the COVID-19 pandemic on the insurance claim system which required customers need to go to the office.This is due to social distancing regulations and requires working from home.Following the issue, insurance companies create applications that make it easier for customers such as insurance claims, payments, and application the insurance.There are 4 types of insurance available, Bebas Rencana Optimal, Sprint Link Smart, Critical Armor and BIPS.These are insurance can protect and cover all types of diseases and all age groups.The aim of this research is to classify customers using the Naï ve Bayes method at PT XYZ and to find out the level of accuracy.The Naï ve Bayes method is to classify probability by adding up the frequencies and combinations of values from a given dataset.Besides that, not only for insurance claims and payments, this application is a place of communication between insurance companies and customers.With this application, it is expected to strengthen customer relationships with insurance companies and make it easier for customers to make transactions and claim insurance.Regarding the accuracy level of product recommendations from 700 data, it was found that accuracy reached more than 92% in recommending types of insurance products to customers.Especially in this era of high technology, which must be utilized by companies in developing the company itself.