Data Mining untuk Nasabah Bank Telemarketing Menggunakan kombinasi Algoritm Naïve Bayes Dan Algoritma Genetik

Abstract

The strategy used for telemarketing by conducting promotional media, this strategy is a marketing method used by banks, in offering products to customers, banks, one of the products that will be offered is time deposits, the bank has difficulty in knowing the obstacles experienced by customers in making a decision to make deposits against the bank, so that later it will have the effect of a financial crisis at the bank. Telemarketing banks must have targets for customers, where customers have the potential to join one of the bank's products, namely deposits by looking at existing customer data.With the existing problems will be overcome by the datamining technique that will be used for this research is the Naïve Bayes algorithm and genetic algorithm which aims to predict the Telemarketing customers' sources sourced from public UCI Repsitory data so that the bank offers a product to the customer right at the target. Naïve Bayes test with experimental results of 86.71% accuracy while cross validation testing using Genetic algorithm produces high accuracy 90.27%, Root proves the prediction of time series data Naïve Bayes method and Genetics produces an accuracy of 90.27%, so it can be concluded that using the Naive Bayes algorithm and Genetics can optimize in predicting Telemarketing client decisions right in the deposit offer.