Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/158059Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Jaya, I Nengah Surati | - |
| dc.contributor.author | Suryandika, Mohammad Fahri | - |
| dc.date.accessioned | 2024-08-21T06:18:39Z | - |
| dc.date.available | 2024-08-21T06:18:39Z | - |
| dc.date.issued | 2024 | - |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/158059 | - |
| dc.description.abstract | Penelitian ini menjelaskan tentang pembangunan algoritma untuk mengidentifikasi sebaran kopi agroforestri dan kopi monokultur di Kabupaten Bandung menggunakan pendekatan pembelajar mesin dengan algoritma pohon keputusan, berbasis citra resolusi tinggi (SPOT-7) dan kondisi bio-geofisik. Penelitian ini bertujuan untuk membangun algoritma pohon keputusan pembelajaran mesin untuk mengidentifikasi sebaran kopi agroforestri dan monokultur di Kabupaten Bandung dengan menggunakan variabel spektral yang diturunkan dari citra SPOT-7 dan peubah-peubah bio-geofisik. Metode pohon keputusan yang diujicobakan menggunakan konsep entropi dengan parameter information gain, gini index dan gain ratio. Kajian ini menemukan bahwa algoritma terbaik untuk identifikasi sebaran spasial kopi agroforestri dan kopi monokultur adalah algoritma decision tree dengan parameter information gain. Kombinasi peubah yang terbaik adalah kombinasi peubah bio-geofisik yaitu slope, jalan, sungai, elevasi dan tutupan lahan, dan peubah spektral, yaitu GARI, SAVI, NRGI, NDWIG, VDVI dan ARVI. Algoritma tersebut menghasilkan akurasi umum (overall accuracy) sebesar (95,12 %) dan kappa accuracy sebesar (0.944). | - |
| dc.description.abstract | This research describes the development of an algorithm to identify the distribution of agroforestry and monoculture coffee in Bandung Regency using a decision tree of machine learning based on a high-resolution SPOT-7 imagery and geo-biophysical factors. The objective of the study is to build a machine learning decision tree algorithm to identify the spatial distribution of agroforestry and monoculture coffee in Bandung Regency by using spectral variables derived from SPOT-7 imagery and geo-biophysical variables. The tested decision tree method employs entropy concepts with parameters including information gain, gini index, and gain ratio. The study found that the best algorithm for identifying the spatial distribution of agroforestry and monoculture coffee is the decision tree algorithm with information gain parameter. The best combination of variables is a combination of geo-biophysical variables, namely slope, road, river, elevation and land cover, and spectral factors, namely GARI, SAVI, NRGI, NDWIG, VDVI and ARVI. This algorithm provided the best performance with an overall accuracy of (95.12%) and a kappa accuracy of (0.944). | - |
| dc.description.sponsorship | null | - |
| dc.language.iso | id | - |
| dc.publisher | IPB University | id |
| dc.title | Identifikasi Sebaran Spasial Kopi Agroforestri dan Monokultur dengan Pendekatan Pohon Keputusan Pembelajaran Mesin Melalui Citra SPOT-7. Studi Kasus: Kabupaten Bandung, Provinsi Jawa Barat. | id |
| dc.title.alternative | Spatial Distribution Identification of Agroforestry and Monoculture Coffee using Decision Tree of Machine Learning Using SPOT-7 Imagery: A Case Study in Bandung Regency, West Java Province. | - |
| dc.type | Skripsi | - |
| dc.subject.keyword | agroforestri | id |
| dc.subject.keyword | pohon keputusan | id |
| dc.subject.keyword | mesin pembelajar | id |
| dc.subject.keyword | SPOT 7 | id |
| Appears in Collections: | UT - Forest Management | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| cover_E1401201089_c5957a2027764dd593b8727a9c849714.pdf | Cover | 1.54 MB | Adobe PDF | View/Open |
| fulltext_E1401201089_6644669d77fe46eb86943e8644fa4799.pdf Restricted Access | Fulltext | 4.08 MB | Adobe PDF | View/Open |
| lampiran_E1401201089_449eb2fe0c4d4aa99cb73eb9557ae6ac.pdf Restricted Access | Lampiran | 301.34 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.