Analisis Akurasi Dari Model Pemetaan Geologi Permukaan Berbasis Machine Learning Menggunakan Algoritma Random Forest Pada PLTP Patuha

Awali, Muhammad Himam (2024) Analisis Akurasi Dari Model Pemetaan Geologi Permukaan Berbasis Machine Learning Menggunakan Algoritma Random Forest Pada PLTP Patuha. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5017201081-Undergraduate_Thesis.pdf] Text
5017201081-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2026.

Download (6MB) | Request a copy

Abstract

Perkembangan peta geologi berperan penting dalam memenuhi kebutuhan eksplorasi. Penerapan supervised machine learning algoritma Random forest dalam integrasi peta disebabkan memiliki ketahanan tinggi terhadap overfitting dan over training. Penelitian ini dilakukan pada lingkungan sumber panas bumi dengan di PLTP Patuha. Tujuan dari Penelitian ini untuk menguji dan mengevaluasi arsitektur dari random forest dari model prediksi berupa peta geologi. Data yang digunakan pada Penelitian ini meliputi Landsat 8, DEMNas, GGMPlus, dan peta geologi penginderaan jarak jauh. Penelitian dimulai dengan pre-processing, processing, dan post-processing. Hasil penelitian ini menunjukkan model prediksi pemetaan geologi permukaan pada lapangan Patuha memiliki akurasi sebesar 82%, split rasio test data sebesar 10:90, dan fitur paling berpengaruh adalah DEMNas dalam membentuk model prediksi dengan nilai kontribusi 21,6%.
================================================================================================================================
The development of geological maps plays an important role in supporting exploration needs. The application of supervised machine learning with the Random forest algorithm is used in map integration because it has resistance to overfitting and overtraining. This research was conducted on the geothermal source environment with the Patuha Geothermal. The purpose of this study is to test and evaluate the architecture of random forests from a prediction model in the form of a geological map. The data used in this study include landsat 8, DEMNas, GGMPlus, and remote sensing geological maps. The research begins with pre-processing, processing, and post-processing. The results of this study show that the prediction model of surface geological mapping in the Patuha field has an accuracy of 82%, a split ratio of 10:90 test data, and the most importance parameter to create predict model is Landsat 8 in forming a prediction model with contribution value is 21,6%.

Item Type: Thesis (Other)
Uncontrolled Keywords: Akurasi Model, Pemetaan Geologi, Model Prediksi, Random forest, Supervised Machine learning
Subjects: G Geography. Anthropology. Recreation > G Geography (General) > G70.5.I4 Remote sensing
G Geography. Anthropology. Recreation > GB Physical geography > GB651 Subsidences (Earth movements)
Q Science > Q Science (General) > Q180 Gravitation.
Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Q Science > QA Mathematics > QA76.9D338 Data integration
Divisions: Faculty of Civil, Planning, and Geo Engineering (CIVPLAN) > Geophysics Engineering > 33201-(S1) Undergraduate Thesis
Depositing User: Muhammad Himam Awali
Date Deposited: 15 Aug 2024 06:18
Last Modified: 15 Aug 2024 06:18
URI: http://repository.its.ac.id/id/eprint/114971

Actions (login required)

View Item View Item