Claim Missing Document
Check
Articles

Found 3 Documents
Search

Comparison of K-Means and DBSCAN Algorithms in Regional Segmentation in Indonesia Based on Mortality and Fertility Indicators Nafandra, Bunga; Sulistiowati, Dwi; Amalita, Nonong
Jurnal MSA (Matematika dan Statistika serta Aplikasinya) Vol 14 No 1 (2026): VOLUME 14 No 1, 2026
Publisher : Universitas Islam Negeri Alauddin Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/msa.v14i1.61961

Abstract

Mortality and fertility indicators can be interpreted through maternal mortality rates, infant mortality rates, under-five mortality rates, child mortality rates, and crude birth rates. Disparities in mortality and fertility between regions indicate differences in health quality. In Indonesia, the maternal mortality rate has declined over the past decade, from 346 to 189 per 100000 live births, but this figure has not yet reached the global target set by the SDGs of 70 per 100000 live births. To reduce these mortality indicators, it is necessary to identify regions with low health quality so that interventions can be more targeted and focused. Therefore, this study aims to form regional segmentation based on mortality and fertility indicators by comparing two clustering algorithms, namely K-Means and DBSCAN. The data used is secondary data from the Indonesian Central Statistics Agency, taken from the 2020 Population Census. The results of the study found that the K-Means algorithm is better than DBSCAN based on the Silhouette Score, Davies-Bouldin Index, and Calinski-Harabasz Index evaluation matrices, with values of 0.485, 0.804, and 36.34, respectively. This algorithm formed two clusters, with western Indonesia tending to have better health quality than eastern Indonesia. Based on the results of this study, it is hoped that the government can provide focused interventions in eastern Indonesia to improve maternal and child health quality and reduce mortality rates.
Regresi Panel dan Pemetaan Local Indicator of Spatial Association dalam Analisis Kemiskinan di Provinsi Bengkulu Nafandra, Bunga; Amalita, Nonong
Imajiner: Jurnal Matematika dan Pendidikan Matematika Vol 8, No 2 (2026): Imajiner: Jurnal Matematika dan Pendidikan Matematika
Publisher : Universitas PGRI Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26877/imajiner.v8i2.26899

Abstract

Kemiskinan tetap menjadi masalah sosial yang menghambat pencapaian kesejahteraan masyarakat. Di Provinsi Bengkulu, persentase penduduk miskin pada periode 2019–2024 lebih tinggi dari rata-rata nasional. Oleh karena itu, penelitian ini bertujuan untuk menganalisis faktor-faktor yang mempengaruhi tingkat kemiskinan dan mengidentifikasi daerah berisiko tinggi di Provinsi Bengkulu melalui pendekatan regresi data panel dan pemetaan Indikator Lokal Asosiasi Spasial (LISA). Hasil penelitian menunjukkan bahwa peningkatan partisipasi angkatan kerja, pertumbuhan produk domestik bruto regional, dan peningkatan indeks pembangunan manusia berkontribusi pada penurunan persentase penduduk miskin di Provinsi Bengkulu. Sementara itu, hasil analisis spasial lokal dengan LISA mengidentifikasi Kabupaten Bengkulu Selatan sebagai klaster tinggi-tinggi atau hotspot, yaitu daerah dengan tingkat kemiskinan tinggi yang dikelilingi oleh daerah dengan kondisi serupa. Berdasarkan temuan ini, diharapkan semua pemangku kepentingan, baik masyarakat maupun pemerintah, dapat berperan dalam meningkatkan kualitas dan kuantitas angkatan kerja, mendorong pertumbuhan ekonomi, dan memperkuat aspek pendidikan dan kesehatan yang mendukung peningkatan HDI. Selain itu, pemerintah perlu memberikan perhatian khusus dan intervensi pada Kabupaten Bengkulu Selatan sebagai daerah hotspot.
Analisis Pengaruh Kelangkaan, Penyajian Produk, dan Motivasi Hedonis terhadap Pembelian Impulsif Secara Online pada Pengguna TikTok Shop Menggunakan Pendekatan Structural Equation Modeling (SEM) syahwa, nurul mulya; Nafandra, Bunga; Putri, Olga Afrilly; Meilisa, Mira; Mukhti, Tessy Octavia
Imajiner: Jurnal Matematika dan Pendidikan Matematika Vol 8, No 2 (2026): Imajiner: Jurnal Matematika dan Pendidikan Matematika
Publisher : Universitas PGRI Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26877/imajiner.v8i2.27033

Abstract

Impulsive online purchasing is a purchasing behavior that occurs spontaneously and unplanned during the online shopping process. Several factors that can influence impulsive online purchasing include scarcity, product presentation, and hedonistic motivation. The increase in impulsive online purchasing behavior cannot be separated from the digital marketing strategies intensively implemented by TikTok Shop. This research seeks to investigate the effects and relationship between scarcity, product presentation, and hedonistic motivation on online impulsive buying among TikTok Shop users. This study applies Structural Equation Modeling (SEM) as the analytical approach since it examines latent variables that are not directly observable, including scarcity, product presentation, and hedonic motivation. SEM is selected because it is capable of generating more robust and comprehensive results compared to simpler techniques such as multiple linear regression. The results demonstrate that scarcity, product presentation, and hedonic motivation significantly influence online impulse buying among TikTok Shop users, with hedonic motivation exerting a particularly strong effect on impulsive purchasing behavior. These three variables were found to contribute positively to impulsive buying behavior, with hedonistic motivation being the factor with the strongest influence. In the structural model, the coefficient values were 0.240 for scarcity, 0.250 for product presentation, and 0.331 for hedonistic motivation.