Perbandingan hasil clustering menggunakan metode k-means, Gaussian mixture models, dan support vector machine pada sektor listrik negara

Suryaputri, Regina Maureen (2024) Perbandingan hasil clustering menggunakan metode k-means, Gaussian mixture models, dan support vector machine pada sektor listrik negara. Bachelor thesis, Universitas Pelita Harapan.

[img]
Preview
Text (Title)
Title.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (250kB) | Preview
[img]
Preview
Text (Abstract)
Abstract.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (587kB) | Preview
[img]
Preview
Text (ToC)
ToC.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (572kB) | Preview
[img]
Preview
Text (Chapter1)
Chapter1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (657kB) | Preview
[img] Text (Chapter2)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (682kB)
[img] Text (Chapter3)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (676kB)
[img] Text (Chapter4)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (685kB)
[img] Text (Chapter5)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (564kB)
[img]
Preview
Text (Bibliography)
Bibliography.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (563kB) | Preview
[img] Text (Appendices)
Appendices.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.

Download (1MB)

Abstract

Memilih metode yang optimal untuk clustering merupakan langkah yang penting, terutama karena clustering umumnya diterapkan dalam analisis sektor tertentu. Hal ini memberikan kemungkinan untuk proyeksi dan pengembangan lebih lanjut dari sektor tersebut. Pada penelitian ini, dilakukan clustering terhadap 117 negara berdasarkan tujuh indeks sektor listrik, dengan menggunakan data pada 3 periode waktu. Penelitian ini membandingkan kinerja dari tiga metode clustering, yaitu k-means, Gaussian mixture models (GMM), dan support vector machine (SVM). Penelitian ini mengevaluasi akurasi hasil clustering menggunakan model SVM yang menunjukkan bahwa SVM mampu memprediksi klaster dengan akurasi berkisar antara 97,44% hingga 100%. Meskipun demikian, akurasi optimal ini tidak konsisten dalam setiap kondisi. Sedangkan hasil clustering yang diperoleh dengan metode k-means dan GMM memiliki hasil klaster yang berbeda. Namun, analisis menunjukkan bahwa GMM adalah metode yang paling optimal dalam melakukan clustering pada data indeks sektor listrik. Analisis kemudian dilakukan terhadap keterkaitan antara indeks sektor listrik dan indeks gross national income (GNI), mengungkapkan beberapa korelasi yang signifikan. Uji korelasi menunjukkan bahwa hubungan antara akses listrik, konsumsi daya listrik, dan kerugian transmisi dengan GNI cukup kuat, sedangkan variabel lainnya menunjukkan korelasi yang lebih rendah. Hasil ini memberikan wawasan tentang kompleksitas hubungan antara sektor listrik dan pertumbuhan ekonomi./Choosing the optimal clustering method is a crucial step, especially because clustering is commonly applied in analyzing specific sectors. This provides the potential for projections and further development within that sector. In this study, clustering was conducted on 117 countries based on seven electricity sector indices, using data from three different periods. The research compares the performance of three clustering methods: k-means, Gaussian mixture models (GMM), and support vector machine (SVM). The study evaluates the accuracy of clustering results using the SVM model, demonstrating that SVM can predict clusters with accuracy ranging from 97.44% to 100%. However, this optimal accuracy is not consistent under every condition. Meanwhile, clustering results obtained with k-means and GMM show different cluster outcomes. Nevertheless, the analysis indicates that GMM is the most optimal method for clustering electricity sector index data. Further analysis examines the relationship between electricity sector indices and gross national income (GNI), revealing some significant correlations. Correlation tests show that the relationship between access to electricity, electricity consumption, and transmission loss with GNI is quite strong, while other variables show lower correlations. These findings provide insights into the complexity of the relationship between the electricity sector and economic growth.

Item Type: Thesis (Bachelor)
Creators:
CreatorsNIMEmail
Suryaputri, Regina MaureenNIM01112200040reginamaureen0306@gmail.com
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorWidjaja, PetrusNIDN0314095901petrus.widjaja@uph.edu
Thesis advisorLukas, SamuelNIDN0331076001samuel.lukas@uph.edu
Uncontrolled Keywords: klaster; k-means; support vector machine; Gaussian mixture models; listrik; klaster negara
Subjects: Q Science > QA Mathematics
Divisions: University Subject > Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Mathematics
Current > Faculty/School - UPH Karawaci > Faculty of Science and Technology > Mathematics
Depositing User: Regina Maureen
Date Deposited: 02 Feb 2024 02:52
Last Modified: 02 Feb 2024 02:52
URI: http://repository.uph.edu/id/eprint/61373

Actions (login required)

View Item View Item