Pemodelan Status Desa di Provinsi Jawa Timur Menggunakan Rare Event Weighted Logistic Regression (RE-WLR)

Candraningtyas, Clarissa Putri (2023) Pemodelan Status Desa di Provinsi Jawa Timur Menggunakan Rare Event Weighted Logistic Regression (RE-WLR). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06211940000003-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Desa merupakan struktur pemerintahan terendah dalam pembangunan nasional yang memiliki kewenangan untuk menjalankan dan mengelola pemerintahannya secara mandiri (otonomi daerah). Dalam rangka melakukan pemantauan pada perkembangan desa, diperlukan sebuah tolak ukur agar dapat mengetahui pembangunan dan pemberdayaan desa telah dilaksanakan tepat sasaran dalam peningkatan kemandirian desa. Pemantauan status desa diimplementasikan melalui Indeks Desa Membangun (IDM) yang diterbitkan oleh Kementerian Desa, Pembangunan Daerah Tertinggal, dan Transmigrasi Tahun 2021 yang menyatakan bahwa Provinsi Jawa Timur menjadi provinsi dengan persentase desa mandiri tertinggi se-Indonesia yaitu 21,32% sekaligus telah dinyatakan terbebas dari desa tertinggal dan desa sangat tertinggal. Penelitian ini menggunakan data 4.438 desa di Provinsi Jawa Timur yang terdiri dari desa berkembang dan desa mandiri. Persentase desa mandiri di Provinsi Jawa Timur sebesar 15,7% sehingga perlu dilakukan penanganan data imbalanced dengan memperhatikan jumlah data yang cukup besar. Oleh karena itu digunakan metode yang cocok untuk data imbalanced berskala besar yaitu Rare Event Weighted Logistic Regression (RE-WLR). Hasil dari penelitian ini menunjukkan dari 17 variabel prediktor yang digunakan terdapat 12 variabel yang berpengaruh signifikan terhadap model adalah keberadaan Taman Bacaan Masyarakat (TBM) (X2), jumlah apotek (X4), keberadaan ruang publik terbuka (X5), tempat buang sampah sebagian besar keluarga (X6), sumber penghasilan utama sebagian besar penduduk desa (X7), jumlah pasar (X8), jumlah usaha bidang kuliner dan penginapan (X9), agen jasa ekspedisi swasta (X11), jumlah bank umum dan BPR (X12), jumlah unit usaha BUMDES (X13), dan jarak tempuh dari kantor kepala desa ke kantor camat (X14), dan sistem peringatan dini bencana alam (X17). Ketepatan model dalam mengklasifikasikan status desa di Provinsi Jawa Timur menghasilkan akurasi sebesar 90,64%, sensitivitas 55,40%, spesifisitas 97,19%, G-Mean 73,38%, dan AUC sebesar 76,29% sehingga termasuk dalam kategori substansial.
===================================================================================================================================
The village has the lowest government structure in national development which has the authority to run and manage its government independently (regional autonomy). To monitor village development, a benchmark is needed so that village development and empowerment have been carried out on target in increasing village independence. Monitoring village status is implemented through the Developing Village Index (IDM) issued by the Ministry of Villages, Development of lagging areas, and Transmigration in 2021 which states that East Java Province is the province with the highest percentage of independent villages in Indonesia, namely 21.32% and has been declared free from underdeveloped and very underdeveloped villages. This research uses data from 4,438 villages in East Java Province, consisting of developing and independent villages. The percentage of independent villages in East Java Province is 15.7%, so data handling is imbalanced, considering a large amount of data. Therefore, the method used is suitable for the data imbalanced large scale, namely Rare Event Weighted Logistic Regression (RE-WLR). The results of this research indicate that of the 17 predictor variables used, 12 variables have a significant effect on the model, namely the existence of Community Reading Parks (TBM) (X2), number of pharmacies (X4), the existence of open public spaces (X5), where most of the family's trash is disposed of (X6), the main source of income for most of the villagers (X7), the number of markets (X8), the number of culinary and lodging businesses (X9), private shipping agency (X11), number of commercial banks and rural banks (X12), number of BUMDES business units (X13), and the distance from the village head's office to the sub-district office (X14), and natural disaster early warning systems (X17). The model's accuracy in classifying village status in East Java Province resulted in an accuracy of 90.64%, a sensitivity of 55.40%, a specificity of 97.19%, a G-Mean of 73.38%, and an AUC of 76.29% so that it is included in the substantial category.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Imbalanced Data, Klasifikasi, Rare Event Weighted Logistic Regression, Status Desa. Imbalanced Data, Classification, Rare Event Weighted Logistic Regression, Village Status.
Subjects:	H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA31.3 Regression. Correlation H Social Sciences > HA Statistics > HA31.7 Estimation Q Science Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Clarissa Putri Candraningtyas
Date Deposited:	09 Aug 2023 06:48
Last Modified:	09 Aug 2023 06:48
URI:	http://repository.its.ac.id/id/eprint/104406

Actions (login required)

View Item