Article Per Year (5 Year)
p-Index From 2021 - 2026
0.659
P-Index
This Author published in this journals
All Journal Faktor Exacta Indonesian Journal of Electrical Engineering and Computer Science KOMMAS: Jurnal Pengabdian Kepada Masyarakat
Kahfi Heryandi Suradiradja
Unknown Affiliation
Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : Faktor Exacta

 1 
Algoritme Machine Learning Multi-Layer Perceptron dan Recurrent Neural Network untuk Prediksi Harga Cabai Merah Besar di Kota Tangerang Kahfi Heryandi Suradiradja
Faktor Exacta Vol 14, No 4 (2021)
Publisher : LPPM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30998/faktorexacta.v14i4.10376

Abstract

Chilli consumption keeps increasing along with the annual population increase in Indonesia. Meanwhile, chilli prices also fluctuate due to rainfall, affecting production and inflation. In the industrial era 4.0, IT support is crucial in various fields including in agriculture such as chilli planting to help stakeholders, both in the economy and agriculture sectors, make decisions based on accurate predictive data support. The study aims to compare the accuracy of two machine learning algorithm models, i.e., Multilayer Perceptron (MLP) and Recurrent Neural Network (RNN), for time-series regression implementable to predict chilli prices in Tangerang City. The experimental method stages include business understanding, data understanding, data preparation, modelling, evaluation, and deployment stages. The required dataset attributes include red chilli prices, date, inflation, and rainfall. This research is expected to contribute to machine learning algorithms to assist stakeholders and to be implemented by information system developers. The research result indicates that the MLP algorithm with the rmsprop optimizer performs better than the RNN with the metric measurement of Loss = 0.0038, MSE = 10271959,0 and MAPE = 3.79%. Suggestions for further research include the urgency to innovate architectural models, either for activation functions, optimizers, or other regression algorithms for better metric measurement results.
Co-Authors Dede Sahrul Bahri Irman Hermadi Luki Abdullah Sitanggang, Imas S. Yulies Herni