Skripsi
IMPLEMENTASI ALGORITMA REGRESI LINEAR UNTUK MEMPREDIKSI PENJUALAN MOTOR DOMESTIK DAN EKSPOR
Motorcycle sales in Indonesia continue to fluctuate in both domestic and export markets, creating the need for sales prediction to support decision-making in the automotive industry. This study aims to predict domestic and export motorcycle sales using the linear regression algorithm supported by the RapidMiner application. The data were obtained from the Indonesian Motorcycle Industry Association (AISI) covering the period from January 2022 to October 2025. The research stages include data collection, dataset normalization, manual calculation of the linear regression formula, and validation through RapidMiner. The results of the study produced the regression equation Y = 67.518 − 0.0327X, indicating a negative relationship between domestic and export sales. This means that every 1-unit increase in domestic sales correlates with a decrease of approximately 0.0327 units in export sales. The prediction results for the next six months show a gradual decline in export sales from 49,009 units to 43,165 units by March 2026. The standard error of the intercept (8.810) and the standard error of the coefficient (0.017204761) indicate stable parameter estimation; however, the F-test value shows Fcalculated < Ftable, signifying that the relationship between variables remains weak. Overall, the linear regression model is capable of providing an initial predictive overview of export sales trends, although model accuracy can be improved by incorporating additional variables such as price, brand, promotion, and consumer purchasing power. Keyword: Linear Regression, Sales Prediction, Motorcycles, Data Mining, RapidMiner, Domestic Sales, Export Sales
| Title | Edition | Language |
|---|---|---|
| PREDIKSI PENGISIAN DAYA POWER SUPPLY DARI SOLAR CELL MENGGUNAKAN ALGORITMA REGRESI LINEAR | id |