<![CDATA[Sales at minimarkets have had ups and downs since the Covid-19 pandemic. In this study, the problem was found, namely the number of products in minimarkets that were not sold or vice versa, there was a demand for products, but they were not available in minimarkets. Data mining can be a solution to solving this problem. This study proposes a hybrid data mining method by combining the K-Means and K-Nearest Neighbors algorithms. The combination of this method works in two stages, namely clustering the products sold which are then continued by classifying the sales of these products. The data used in this study is data for toiletries and washing products as many as 20 products. From the results of research conducted, there are 14 products that have many enthusiasts from 20 products. Of the 14 products used as training data, a sales classification was carried out for 1 new product with stock criteria 200, sold 100, and price Rp. 10,000. From the test results, the new product classification is High Sales with accuracy in the classification reaching 85.7143%. The use of a hybrid method between the K-Means and K-Nearest Neighbors algorithms has a significant influence in determining the classification results. For further research, it is recommended to have a prediction process from the classification results that have been found, so that they have a greater influence on the use of this hybrid data mining method.]]>
Copyrights © 2022
