Jurnal Teknik Informatika (JUTIF)
Vol. 4 No. 2 (2023): JUTIF Volume 4, Number 2, April 2023

DECISION TREE SIMPLIFICATION THROUGH FEATURE SELECTION APPROACH IN SELECTING FISH FEED SELLERS

Esmi Nur Fitri (Jurusan Sistem Informasi, Fakultas Ilmu Komputer, Universitas Dian Nuswantoro, Indonesia)
Sri Winarno (Jurusan Sistem Informasi, Fakultas Ilmu Komputer, Universitas Dian Nuswantoro, Indonesia)
Fikri Budiman (Jurusan Teknik Informatika, Fakultas Ilmu Komputer, Universitas Dian Nuswantoro, Indonesia)
Asih Rohmani (Jurusan Sistem Informasi, Fakultas Ilmu Komputer, Universitas Dian Nuswantoro, Indonesia)
Junta Zeniarja (Jurusan Teknik Informatika, Fakultas Ilmu Komputer, Universitas Dian Nuswantoro, Indonesia)
Edi Sugiarto (Jurusan Teknik Informatika, Fakultas Ilmu Komputer, Universitas Dian Nuswantoro, Indonesia)



Article Info

Publish Date
23 Mar 2023

Abstract

Feed is a crucial variable because it can determine the success of fish farming. Breeders can use two types of artificial feed, namely alternative feed and pellets. Many cultivators need pellets as the main consumption for the fish they are cultivating because the pellets contain a composition that has been adjusted to their needs based on the type and age of the fish. However, currently, cultivators are facing a problem, namely the high price of fish pellets on the market. Therefore, an analysis of the classification of the selection of fish feed sellers is needed that is adjusted to several criteria like the number of types of feed, price, order, delivery, and availability of discounts. This study conducted a classification analysis of simplification of characteristics in selecting fish feed sellers in Kendal Regency that would then be compared with a model without feature selection by utilizing the Decision Tree C4.5 method. The results of this study are the decision tree with the best performance where C4.5 with the application of the selected feature has an accuracy value of 92%, while C4.5 without the selection feature has an accuracy of 86.8%. The results of this study indicate that the C4.5 method with the application of selection features is better than C4.5 without selection features so that it can be applied to the selection of freshwater fish feed sellers in Kendal Regency.

Copyrights © 2023






Journal Info

Abbrev

jurnal

Publisher

Subject

Computer Science & IT

Description

Jurnal Teknik Informatika (JUTIF) is an Indonesian national journal, publishes high-quality research papers in the broad field of Informatics, Information Systems and Computer Science, which encompasses software engineering, information system development, computer systems, computer network, ...