Main Article Content

Abstract

In carrying out a business that operates in the field of selling goods, predictions are very necessary. Predictions are very necessary because making predictions can help predict what will happen in the future so that the risks that will occur can be minimized as small as possible. This research focuses on sales predictions in 2023 for the HT Motorola XiR C2660 brand using the SVR (support vector regression) algorithm and using linear type kernel parameter testing and c(cost) with a test value of 0.1. To obtain parameter types and parameter test values, use the GridSearchCV technique. Apart from that, this research uses error value testing with mean absolute percentage error (MAPE). So the results of the HT Motorola XiR C2660 sales prediction for 2023 with the SVR algorithm were obtained. In 2023, 209 units were sold with an error value of 11.23%, which means the forecasting ability with the SVR algorithm is good.

Keywords

MAPE, prediction, sale, SVR

Article Details

How to Cite
Wildwina, & Ryan Putranda Kristianto. (2024). Prediksi Penjualan HT Motorola XiR C2660 Menggunakan Algoritma Support Vector Regression (Studi Kasus: CV. Alfacoms). Jurnal Teknik Informatika Dan Komputer, 3(1), 11–16. https://doi.org/10.22236/jutikom.v3i1.13836

References

Read More