PENERAPAN ALGORITMA K-NEAREST NEIGHBOR UNTUK MEMPREDIKSI PENJUALAN SEPEDA MOTOR TERLARIS PADA PT DAYA ANUGRAH MANDIRI

Abstract

Motorcycles are an option for everyone as a means of transportation because they are cheap and can be used for a long time. The increase in the price of consumer goods or daily necessities such as the increase in fuel prices, even on religious holidays (eid), or often Also known as fluctuation (seasonal), the problem that often occurs in making sales plans is when sales predictions are made. If the company is too small, the company will run out of goods and vice versa. The purpose of this research is that it is hoped that this research can help companies to make decisions in providing stock. Processing of motorcycle sales data as much as 170 and 3 attributes that exist in the data selection using the K-Nearest Neighbor algorithm method produces predictions of sales of Honda motorcycles with the metric type more in demand by consumers than the sport and CUB types. This study uses the K-Nearest Neighbord algorithm with an accuracy value of 97.65%.