PEMODELAN EXTREME LEARNING MACHINE (ELM) DENGAN OPTIMASI MODIFIED ARTIFICIAL BEE COLONY UNTUK MERAMALKAN HARGA BATU BARA DUNIA

Eka, Kurniawati (2022) PEMODELAN EXTREME LEARNING MACHINE (ELM) DENGAN OPTIMASI MODIFIED ARTIFICIAL BEE COLONY UNTUK MERAMALKAN HARGA BATU BARA DUNIA. Sarjana / Sarjana Terapan (S1/D4) thesis, Universitas Muhammadiyah Semarang.

[img]
Preview
Text
COVER.pdf

Download (445kB) | Preview
[img]
Preview
Text
ABSTRAK.pdf

Download (432kB) | Preview
[img]
Preview
Text
BAB I.pdf

Download (535kB) | Preview
[img]
Preview
Text
BAB II.pdf

Download (932kB) | Preview
[img] Text
BAB III.pdf
Restricted to Repository staff only

Download (540kB) | Request a copy
[img] Text
BAB IV.pdf
Restricted to Repository staff only

Download (1MB) | Request a copy
[img]
Preview
Text
BAB V.pdf

Download (481kB) | Preview
[img]
Preview
Text
DAFTAR PUSTAKA.pdf

Download (500kB) | Preview
[img] Text
LAMPIRAN.pdf
Restricted to Repository staff only

Download (995kB) | Request a copy
[img] Text
FULL TEXT.pdf
Restricted to Repository staff only

Download (3MB) | Request a copy
Official URL: http://repository.unimus.ac.id

Abstract

ABSTRAK Kurniawati, Eka, 2022, Pemodelan Extreme Learning Machine (ELM) dengan Optimasi Modified Artificial Bee Colony untuk Meramalkan Harga Batu Bara Dunia. Skripsi, Program Studi Statistika, Universitas Muhammadiyah Semarang. Pembimbing: I. Tiani Wahyu Utami, S.Si., M.Si., II. Dr. Rochdi Wasono, M.Si. Batu bara merupakan salah satu sumber energi yang dibutuhkan dunia saat ini. Pergerakan harga batu bara dunia yang tidak menentu dan cenderung dinamis menyebabkan peramalan harga batu bara dunia perlu dilakukan guna mengetahui trend harga untuk melakukan perencanaan tambang dalam jangka waktu panjang. Salah satu metode jaringan syaraf tiruan yang memiliki learning speed yang cepat dan tingkat kesalahan yang rendah adalah Extreme Learning Machine (ELM). Namun, ELM memiliki kekurangan dalam hal pemilihan bobot masukan dan bias yang dilakukan secara acak. Kekurangan tersebut dapat diatasi dengan bantuan dari metode optimasi Modified Artificial Bee Colony yang digunakan untuk menghasilkan bobot masukan dan bias paling optimal pada metode ELM. Tujuan penelitian ini yaitu untuk menerapkan model, mendapatkan hasil peramalan harga batu bara dunia, dan tingkat akurasi peramalan menggunakan ELM dengan optimasi Modified Artificial Bee Colony. Kriteria pemilihan model terbaik menggunakan nilai MAPE. Berdasarkan hasil analisis, diperoleh kombinasi terbaik dari model dengan 5 input neuron, 10 hidden neuron, 8 populasi bee, dan 4 iterasi maksimum. Kombinasi tersebut menghasilkan nilai MAPE terkecil yaitu 1.3261% dan akurasi peramalan sebesar 98.6739%. Kata Kunci : Extreme Learning Machine, Harga Batu Bara, Modified Artificial Bee Colony, Peramalan   ABSTRACT Kurniawati, Eka, 2022, Extreme Learning Machine (ELM) Modeling with Optimization of Modified Artificial Bee Colony to Predict World Coal Prices. Thesis, Statistics Study Program, University of Muhammadiyah Semarang. Supervisor: I. Tiani Wahyu Utami, S.Si., M.Si., II. Dr. Rochdi Wasono, M.Si. Coal is one of the energy sources that the world needs today. The erratic and dynamic movement of world coal prices causes world coal price forecasting to be carried out to determine price trends for long-term mine planning. One of the artificial neural network methods with a fast learning speed and a low error rate is Extreme Learning Machine (ELM). However, ELM has drawbacks regarding a random selection of input weights and bias. These shortcomings can be overcome with the help of the Modified Artificial Bee Colony optimization method, which is used to produce the most optimal input weight and bias in the ELM method. The purpose of this study is to apply the model, obtain the results of world coal price forecasting, and the level of forecasting accuracy using ELM with Modified Artificial Bee Colony optimization. The criteria for selecting the best model use the MAPE value. Based on the analysis results, we obtained the best combination of the model with 5 input neurons, 10 hidden neurons, 8 bee populations, and 4 maximum iterations. This combination produces the smallest MAPE value of 1.3261% and forecasting accuracy of 98.6739%. Keywords : Coal Price, Extreme Learning Machine, Forecasting, Modified Artificial Bee Colony

Item Type: Thesis (Sarjana / Sarjana Terapan (S1/D4) )
Call Number: 020/Statistika/IX/2022
Subjects: L Education > Statistics
Divisions: Faculty of Agricultural Science and Technology > S1 Statistics
Depositing User: perpus unimus
URI: http://repository.unimus.ac.id/id/eprint/5994

Actions (login required)

View Item View Item