Implementation of K-means algorithm in data analysis

Abstract

Some large companies have difficulty in providing products even though the products are still available in the warehouse. Based on these problems, a solution is needed in managing cosmetic products and can find the right strategy so that it can increase business in the field of sales and improve sales services by using algorithms in data mining that can overcome these problems, such as clustering techniques that use the K-means clustering algorithm as a way to measure proximity data between cosmetic products based on transactions that have occurred. The specialty of the analysis of the management of cosmetic products in this study is that it produces data on products that are not sold enough so that it can provide prevention so that the accumulation of these products does not occur. The use of K-means clustering also makes it easier to collect cosmetic sales transaction data, can solve problems in classifying cosmetic product sales transaction data and find out which products should be in cosmetics stock so as to increase sales profits