IMPLEMENTASI GAUSSIAN MIXTURE MODEL PADA APLIKASI WEB XPORTID UNTUK KLASTERISASI KOMODITAS EKSPOR INDONESIA BERDASARKAN BENUA TUJUAN

Lisanthoni, Angela (2025) IMPLEMENTASI GAUSSIAN MIXTURE MODEL PADA APLIKASI WEB XPORTID UNTUK KLASTERISASI KOMODITAS EKSPOR INDONESIA BERDASARKAN BENUA TUJUAN. Undergraduate thesis, Universitas Pembangunan Nasional Veteran Jawa Timur.

	Text (Cover) 21083010032_Cover.pdf Download (1MB)
	Text (Bab 1) 21083010032_Bab I.pdf Download (91kB)
	Text (Bab 2) 21083010032_Bab II.pdf Restricted to Registered users only until 31 January 2028. Download (598kB)
	Text (Bab 3) 21083010032_Bab III.pdf Restricted to Registered users only until 31 January 2028. Download (667kB)
	Text (Bab 4) 21083010032_Bab IV.pdf Restricted to Registered users only until 31 January 2028. Download (3MB)
	Text (Bab 5) 21083010032_Bab V.pdf Download (13kB)
	Text (Daftar Pustaka) 21083010032_Daftar Pustaka.pdf Download (101kB)
	Text (Lampiran) 21083010032_Lampiran.pdf Restricted to Registered users only until 31 January 2028. Download (606kB)

Abstract

The increase in Indonesia's exports is crucial for strengthening foreign exchange reserves and driving economic growth. However, in 2023, Indonesia's export value decreased by 11%, indicating that export performance is not optimal. One of the causes of the decline is the limited selection of export market destinations. Therefore, this study aims to analyze Indonesia's trade patterns to identify the most potential markets for increasing the efficiency of export marketing strategies and designing a website that presents clustering results and data according to user input. The method used is clustering with a Gaussian Mixture Model (GMM) approach which implements the Expectation-Maximation (EM) algorithm to identify the best cluster parameters. GMM has advantages in terms of flexibility and uses probability so that the results obtained are more accurate. This study goes through several stages: data collection, data preprocessing, exploratory data analysis, GMM model construction, model evaluation using silhouette score, and the development of a Flask-based website. The clustering results achieved an average silhouette score of 0.8185, indicating a strong structure. Clustering results by continent show that Asia has a value of 0.7035 (good structure) with five clusters, America gets a value of 0.8534 (strong structure) with three clusters, Africa has a value of 0.8165 (strong structure) with three clusters, Australia reaches a value of 0.8540 (strong structure) with three clusters, and Europe gets a value of 0.8654 (strong structure) with three clusters. The clustering results in overall show countries with high export potential include Malaysia, the Philippines, South Korea, Brazil, Mexico, New Zealand, and Spain. In addition, the XportID website that was developed allows users to input data and get the results of export commodity clustering. Keywords: Export, Clustering, Gaussian Mixture Model (GMM), Silhouette Score, Flask

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Trimono, Trimono	NIDN0008099501	trimono.stat@upnjatim.ac.id
Thesis advisor	Prasetya, Dwi Arman	NIDN0005128001	arman.prasetya.sada@upnjatim.ac.id

Subjects:

H Social Sciences > HA Statistics
Q Science > QA Mathematics > QA76.6 Computer Programming

Divisions:

Faculty of Computer Science > Departemen of Data Science

Depositing User:

Angela Lisanthoni

Date Deposited:

31 Jan 2025 08:17

Last Modified:

31 Jan 2025 08:17

URI:

https://repository.upnjatim.ac.id/id/eprint/34454