Penerapan Algoritma Naive Bayes Untuk Menentukan Klasifikasi Produk Terlaris Pada Penjualan Pulsa

Abstract

This research is motivated by the progress of the development of communication technology and information is very fast and increasingly cheap so that makes the community for mobile phone pulses become a mandatory requirement at the presente time. Of various pulse products available at the counter RA Cell Pulses Tekomsel, Pulses XL, Pulses Indosat, and Pulses 3 the autors classify as bestseller and non-sellers. The goal is to find out the implementation of data mining using the Naive Bayes algorithm in determining the classification of best-selling products and the result of the accuracy of the data in the sales of pulses. By collecting 600 data into 480 training data and 120 testing data. Data mining is a form of extracting data in classifying a large amount of data, using the RapidMiner application and the Naive Bayes algorithm is a classification method that is widely used because of its simple and high accuracy in classifying data. Based on the result of researh that has been done, the type of product that is most restricted to the sale of pulses by product name is Telkomsel Pulses. The level of classification accuracy with the naive Bayes methodproduces an accuracy value of 97,50%, a precision value of 100%, and a recall value of 93,48% so the Naive Bayes method is good method in this study.

Keywords :Pulses, Classification, Data Mining, Naive Bayes