Penerapan Metode Rare Event Weighted Logistic Regression untuk Kasus Data Imbalanced (Studi Kasus : Jenis Perceraian di Kabupaten Bojonegoro)

Merdiko, Cinde Pristi Kurnia (2021) Penerapan Metode Rare Event Weighted Logistic Regression untuk Kasus Data Imbalanced (Studi Kasus : Jenis Perceraian di Kabupaten Bojonegoro). Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

Text 06211740000015-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (1MB) | Request a copy

Abstract

Jawa Timur menjadi provinsi penyumbang angka perceraian tertinggi di Indonesia beberapa tahun terakhir. Pada tahun 2019, jumlah kasus perceraian di Jawa Timur meningkat sebesar 6.052 kasus dari tahun sebelumnya dan menyumbang 21,6% perceraian di Indonesia. Kabupaten Bojonegoro menjadi salah satu penyumbang tingginya angka perceraian di Jawa Timur. Perceraian menurut kasusnya dibagi menjadi dua yaitu cerai gugat dan cerai talak sehingga salah satu metode yang bisa digunakan untuk mengidentifikasi faktor signifikan penyebab perceraian adalah metode regresi logistik. Akan tetapi, kasus perceraian di Kabupaten Bojonegoro tahun 2020 merupakan data imbalanced dengan perbandingan cerai gugat dan cerai talak adalah 71,4% dan 28,6% sehingga pada penelitian ini digunakan metode pengembangan regresi logistik untuk data imbalanced yaitu Rare Event Weighted Logistic Regression (RE-WLR). Hasil penelitian ini menunjukkan bahwa usia penggugat/pemohon dan tergugat/termohon ketika menikah, pekerjaan penggugat/permohon dan tergugat/termohon, serta sebab perceraian berpengaruh signifikan terhadap jenis perceraian di Kabupaten Bojonegoro dengan nilai ketepatan klasifikasi AUC yang dihasilkan sebesar 0,852 dan termasuk dalam Good Classification.
================================================================================================
East Java has become the province that contributes the highest divorce rate in Indonesia in recent years. In 2019, the number of divorce cases in East Java increased by 6,052 cases from the previous year and contributed for 21.6% of divorces in Indonesia. Bojonegoro Regency is one of the contributors to the high divorce rate in East Java. Divorce according to the case is divided into two, such as divorce applied by wife (gugat divorce) and divorce applied by husband (talak divorce), so one of the method that can be used to identify significant factors causing divorce is the logistic regression method. However, the divorce case in Bojonegoro Regency in 2020 is imbalanced data with ratio of gugat divorce and talak divorce is 71.4% and 28.6%, so this study uses a logistic regression development method for imbalanced data, namely Rare Event Weighted Logistic Regression (RE-WLR). The results of this study indicate that the age at married, the types of occupation, and the cause of divorce have a significant effect on the types of divorce in Bojonegoro Regency with the classification accuracy (AUC) value is 0.852 and included in the Good Classification.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	jenis perceraian, rare event weighted logistic regression, types of divorce
Subjects:	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:	Cinde Pristi Kurnia Merdiko
Date Deposited:	31 Aug 2021 17:46
Last Modified:	31 Aug 2021 17:46
URI:	http://repository.its.ac.id/id/eprint/91350

Actions (login required)

View Item