Page 36 - Proceeding The 2nd International Seminar of Science and Technology : Accelerating Sustainable Innovation Towards Society 5.0
P. 36
nd
The 2 International Seminar of Science and Technology
“Accelerating Sustainable innovation towards Society 5.0”
ISST 2022 FST UT 2022
Universitas Terbuka
SVM, Decision Tree, Random Forest and Linear Regression methods.
The results of this research indicate that Random Forest is the most
optimal method with a MAPE value of 12%. Similar research was also
carried out by Rianto and Yunis [9] to forecast the number of new
students using the Random Forest method. The result of this research
is that the MSE and MAE values are 0.02%, with an accuracy of
99.8%. Based on several previous studies, the Random Forest and
Single Exponential Smoothing methods are more optimal than the
comparison methods, so the purpose of this study is to compare the
performance of these two methods in forecasting the demand for gold
for Indonesian jewellery.
2 METHODOLOGY
The data used in this study is secondary data, namely data on the
demand for gold for Indonesian jewellery for the period 2010 to 2021.
The data is obtained from the World Gold Council (WGC) website
which is accessed via https://gold.org. This study uses 2 variables,
namely the time variable (quarterly) and the gold jewellery demand
variable (tons). Analysis of the data in this study using the Random
Forest and Single Exponential Smoothing methods using the R
software.
2.1 Random Forest
Random Forest is a method that consists of a structured set of trees,
each of which casts a unit of votes for the class and the results
obtained according to the most decisions. The basic technique of
Random Forest is the decision tree. Random Forest is a set of
decision trees that are used for classifying and predicting data by
entering input into the roots at the top and then down to the leaves at
the bottom [10]. The results of the Random Forest analysis for
classification are the mode of each tree of the built forest, while the
prediction results are obtained from the average value of each tree
[11].
ISST 2022 – FST Universitas Terbuka, Indonesia 15
International Seminar of Science and Technology “Accelerating Sustainable
Towards Society 5.0