Jurnal SINTA 2
Konsep Decision Tree Reptree untuk Melakukan Optimasi Rule dalam Fuzzy Inference System Tsukamoto
Jurnal Teknologi Informasi dan Ilmu Komputer (JTIIK) · Vol. 9 No. 3 (2022) · pp. 513–522
- ISSN
- 2528-6579, 2355-7699
- Published
- 2022-06-20
- Publisher
- Fakultas Ilmu Komputer, Universitas Brawijaya
- Language
- id
Abstract
Penelitian ini menjelaskan tentang decision tree REPTree dalam membuat suatu rule yang terbentuk dari produksi minyak kelapa sawit di PT Tapiana Nadenggan, yang dipengaruhi oleh faktor banyaknya kelapa sawit, permintaan yang ada, serta persediaan yang tersedia. Konsep dari decision tree REPTree adalah konsep awal dari decision tree J48 yang kemudian mengalami pemangkasan kembali, sehingg rule yang yang terbentuk lebih minimal dan praktis. Rule yang minimal dan praktis belum tentu dapat dikatakan terbaik, untuk membuktikan hal itu perlu adanya uji coba dan pembuktian. Pembuktian yang dilakukan dalam penelitian ini salah satunya dengan menggunakan perbandingan decision tree J48 dan Random Tree dengan tujuan untuk mengetahui optimasi rule yang terbentuk dengan menggunakan metode fuzzy inference system Tsukamoto, setelah dihitung bahwa decision tree REPTree mempunyai Average Forecasting Error Rate (AFER) yang lebih kecil sebesar 23,17% dengan nilai kebenaran 76,83%, sedangkan J48 memiliki tingkat error sebesar 24,96%, dengan nilai kebenaran 75,04%, sementara Random Tree memiliki tingkat error sebesar 36,51%, dengan nilai kebenaran 63,49% pada kasus prediksi produksi minyak kelapa sawit di PT Tapiana Nadenggan. AbstractThis research explains about REPTree's decision tree in making a rule that is formed from the production of palm oil in PT Tapiana Nadenggan, which is influenced by factors of the amount of palm oil, existing demand, and available supplies. The concept of the REPTree decision tree is the initial concept of the J48 decision tree which then experiences pruning, so that the rules formed are more minimal and practical. A minimum and practical rule may not be the best, to prove that there is a need for trials and proofs. Proof carried out in this research is one of them by using a comparison of decision trees J48 and Random Tree with the aim to find out the optimization of rules formed using the Tsukamoto system's fuzzy inference method, after calculating that the REPTree decision tree has a more average Forecasting Error Rate (AFER) error tree small of 23.17% with a truth value of 76.83%, while J48 has an error rate of 24.96%, with a truth value of 75.04%, while Random Tree has an error rate of 36.51%, with a truth value of 63, 49% in the case of prediction of palm oil production at PT Tapiana Nadenggan.
Collaboration network of the authors
12 authors · 13 co-authorship links · 6 across different organisations
Hover or tap a name to highlight its co-authors. Double-click to open the author's page.
- Sunan Kalijaga State Islamic University Yogyakarta
- Universitas Teknologi Yogyakarta
- Universitas Gadjah Mada
- AGH University of Krakow
- Affiliation unknown
- Different organisations
- Same organisation
- Affiliation unknown
- Thicker line = more joint articles · bigger circle = more articles
Show as table
| Co-author | Organisation | Joint articles |
|---|---|---|
| Riolandi Akbar | Sunan Kalijaga State Islamic University Yogyakarta | 2 |
| Subanar Subanar | Universitas Gadjah Mada | 1 |
| Toni Efendi | Sunan Kalijaga State Islamic University Yogyakarta | 1 |
| Ahmad Mustafid | Sunan Kalijaga State Islamic University Yogyakarta | 1 |
| Enny Itje Sela | Universitas Teknologi Yogyakarta | 1 |
| Doni Kurniawan | Universitas Teknologi Yogyakarta | 1 |
| Yudi Istianto | Sunan Kalijaga State Islamic University Yogyakarta | 1 |
| Wisnu Dwi Nugroho | Universitas Teknologi Yogyakarta | 1 |
| Shoffan Saifullah | AGH University of Krakow | 1 |
| Yusril Iza Fajarendra | Affiliation unknown | 1 |
Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.