A Decision Tree Combination from REP-Random in Creating a Rule to Determine Palm Oil Predictions Using FIS Tsukamoto

Tundo, Tundo

2023

Abstract

This research explains various decision trees and updates to decision trees by carrying out combination scenarios of decision trees that are used in taking a rule that is formed from palm oil production at PT Tapiana Nadenggan which is influenced by factors such as the amount of palm oil, existing demand, and available supplies. The decision trees used in this research, particularly REP and Random, were continued with the combination scenario concept of these decision trees by giving the combined name of the decision trees, specifically REP-Random. The purpose of the decision tree combination is an updated idea related to the combination concept to recognize a performance process from the selected decision tree combination, which in accuracy can exceed the results of the decision tree used or even the results are much worse. To answer this, it is necessary to have a method that is used to find out the performance comparison of decision trees and combinations. The method used is the Tsukamoto Fuzzy Inference System, to determine palm oil production output based on forecasting. After carrying out the forecasting process, the REP-Random combination decision tree is the most recommended alternative, producing the greatest accuracy with a value of 91.18%, followed by REP at 85.68%, and finally Random at 74.14%. The results that have been carried out mean that the decision tree combination scenario can be used to carry out forecasting in creating automatic rules based on the trials that have been carried out.

Paper Citation

in Harvard Style

Tundo. (2023). A Decision Tree Combination from REP-Random in Creating a Rule to Determine Palm Oil Predictions Using FIS Tsukamoto. In Proceedings of the 4th International Seminar and Call for Paper - Volume 1: ISCP UTA '45 JAKARTA; ISBN 978-989-758-691-0, SciTePress, pages 416-421. DOI: 10.5220/0012583800003821

in Bibtex Style

@conference{iscp uta '45 jakarta23,
author={Tundo},
title={A Decision Tree Combination from REP-Random in Creating a Rule to Determine Palm Oil Predictions Using FIS Tsukamoto},
booktitle={Proceedings of the 4th International Seminar and Call for Paper - Volume 1: ISCP UTA '45 JAKARTA},
year={2023},
pages={416-421},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012583800003821},
isbn={978-989-758-691-0},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 4th International Seminar and Call for Paper - Volume 1: ISCP UTA '45 JAKARTA
TI - A Decision Tree Combination from REP-Random in Creating a Rule to Determine Palm Oil Predictions Using FIS Tsukamoto
SN - 978-989-758-691-0
AU - Tundo.
PY - 2023
SP - 416
EP - 421
DO - 10.5220/0012583800003821
PB - SciTePress