| Autor: |
Henderi, H., Ramli, R., Cynthia, Eka Pandu, Sarbaini, S., Muttakin, Fitriani, Nazaruddin, N., Ismanto, Edi, Windarto, Agus Perdana, Mesran, M., Chin, Jacky, Kurniawan, Andi |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2024, Vol. 3065 Issue 1, p1-6, 6p |
| Abstrakt: |
The purpose of this study is to combine Backpropagation and Particle Swarm Optimization (PSO) methods in order to increase the preference for medicinal plants based on their country of origin. The method used in this research is backpropagation followed by particle swarm optimization (PSO). The data for this study were obtained from the Indonesian Central Statistics Agency's (BPS) official website, which can be accessed at the following URL: https://bps.go.id. The data for this study comes from the export of medicinal plants by destination country from 2016 to 2019. The data is processed using the x-validation and Optimize Weights (PSO) features of RapidMiner software version 5.3. Beginning with backpropagation to determine the optimal architecture and Root Mean Square Error (RMSE) results, and then with Particle Swarm Optimization to validate the backpropagation results (PSO). As indicated by the Root Mean Square Error (RMSE) value, the best architectural model was obtained, namely the 2-11-1 model with an RMSE of 0.200 +/-0.000. The combination of Backpropagation and Particle Swarm Optimization (PSO) results in a Root Mean Square Error (RMSE) of 0.217 +/-0.000 (better: 0.017 +/-0.000). [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
