Article Per Year (5 Year)
p-Index From 2019 - 2024
0.23
P-Index
This Author published in this journals
All Journal Nusantara Journal of Computers and its Applications
Hozairi Hozairi
Teknik Informatika, Universitas Islam Madura
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Engineering Other

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
IMPLEMENTATION OF ARTIFICIAL NEURAL NETWORK AND RECURRENT NEURAL NETWORK METHODS TO PREDICT THE AMOUNT OF SALT PRODUCTION Miftahul Walid; Dini Fajariyah; Hozairi Hozairi; Budi Satria
NJCA (Nusantara Journal of Computers and Its Applications) Vol 8, No 1 (2023): June 2023
Publisher : Computer Society of Nahdlatul Ulama (CSNU) Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36564/njca.v8i1.314

Abstract

Sumenep is one of the salt-producing regencies in Madura with 27 sub-districts where 11 sub-districts are salt producers which have a total area of 2,077.12 ha of ponds. Generally, people only cultivate salt in certain months because this salt production can only be done and depends on several factors, such as weather and land area. From the existing problems, this research was conducted using a Deep Learning approach, namely Artificial Neural Network (ANN) and Simple Recurrent Neural Network (SimpleRNN) to predict the amount of salt production. Weather data as input and salt production data as output taken from the last 6 years (2017-2022). The accuracy value in model training was used as a comparison to make predictions. the process of dividing training and testing data was also carried out with a ratio of 80%:20%. Furthermore, both methods was given 6 trainings each, so that the training of the two methods produces a different accuracy value. The ANN model produces an accuracy value of 53% and 71% for Simple RNN. Based on the resulting accuracy value, this base cased study is suitable for using the SimpleRNN algorithm model compared to ANN, provided that the amount of data used is large-scale
Co-Authors Budi Satria Dini Fajariyah Walid, Miftahul