Page 40 - Proceeding The 2nd International Seminar of Science and Technology : Accelerating Sustainable Innovation Towards Society 5.0
P. 40

nd
               The 2  International Seminar of Science and Technology
               “Accelerating Sustainable innovation towards Society 5.0”
               ISST 2022 FST UT 2022
               Universitas Terbuka
                    Table 2. ntree and error values for Random Forest method














               Based on Table. 2. The smallest MAPE, RMSE and MSE values are
               12.866, 1.287 and 1.656 respectively using a ntree value of 50. With
               an  error  value  below  15%,  it  is  included  in  the  good  category.
               Therefore,  the  Random  Forest  method  can  be  used  to  predict  the
               demand for gold in Indonesian jewellery. The following is a graph of
               forecasting results using  the Random Forest method using  a ntree
               value of 50.
               Based on the graph in Figure. 2 The data from the forecasting results
               are quite good in following the data pattern or not much different from
               the  actual  data  pattern.  The  graph  also  shows  the  results  of
               forecasting for each quarter in quartile IV/2021 to quartile IV/2022. As
               seen in the graph, the demand for gold for Indonesian state jewellery
               in the period IV/2021 to the next IV/2022 tends to decline.



















               ISST 2022 – FST Universitas Terbuka, Indonesia             19
               International Seminar of Science and Technology “Accelerating Sustainable
               Towards Society 5.0
   35   36   37   38   39   40   41   42   43   44   45