Analisis Pemodelan Demografi Status Kerja di Kota Blitar dengan Algoritma C4.5 dan Regresi Logistik Ordinal

Limara, Tianta (2024) Analisis Pemodelan Demografi Status Kerja di Kota Blitar dengan Algoritma C4.5 dan Regresi Logistik Ordinal. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5006201080-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2026. Download (3MB) | Request a copy

Abstract

Indonesia merupakan negara berkembang yang masih mengalami permasalahan sosial ekonomi berupa pengangguran. Fenomena pengangguran ini justru mengalami peningkatan setelah terjadinya pandemi Covid-19 yang terjadi di Indonesia karena banyak pekerja yang mengalami PHK akibat pertumbuhan ekonomi negara yang melambat. Pemerintah berupaya menangani masalah pengangguran akibat PHK ini dengan membentuk program Jaminan Kehilangan Pekerjaan (JKP) yang dikelola oleh BPJS Ketenagakerjaan sebagai produk Asuransi Pengangguran di Indonesia. Pengembangan dan evaluasi produk asuransi ini memerlukan penilaian aktuaria yang mencakup pemodelan demografi sebagai tinjauan dalam menggambarkan risiko ketenagakerjaan regional untuk memastikan perlakuan yang adil untuk semua individu. Data demografi di suatu wilayah yang mencakup komposisi tenaga dapat menjadi dasar pemodelan demografi, khususnya dalam konteks program Asuransi Pengangguran. Hampir di setiap kota atau kabupaten di Indonesia, termasuk di Kota Blitar, masih mengalami masalah pengangguran sehingga perlu dilakukan analisis klasifikasi status kerja penduduk di Kota Blitar yang mampu memberikan gambaran kondisi ketenagakerjaan di wilayah Kota Blitar. Penelitian melakukan analisis klasifikasi status kerja dengan model machine learning terhadap data penduduk di Kota Blitar yang dikelompokkan menjadi tiga kategori, yaitu full employed, under employed, dan unemployed. Analisis ini akan menggunakan model machine learning dengan metode Decision Tree Algoritma C4.5 dan Regresi Logistik Ordinal untuk mengetahui model klasifikasi demografi ketenagakerjaan terbaik melalui perbandingan tingkat akurasi model yang terbentuk. Dalam memprediksikan kategori status kerja, model yang terbentuk dari metode Decision Tree Algoritma C4.5 menghasilkan nilai akurasi sebesar 78,18%. Di sisi lain, model yang terbentuk dari metode Regresi Logistik Ordinal menghasilkan nilai akurasi sebesar 79,49%. Berdasarkan hasil tersebut, metode Regresi Logistik Ordinal merupakan metode yang menghasilkan model lebih baik dibandingkan Decision Tree Algoritma C4.5 dalam melakukan prediksi klasifikasi kategori status kerja di Kota Blitar.
=================================================================================================================================
Indonesia is a developing country that still experiences socioeconomic problems in the form of unemployment. This unemployment phenomenon actually increased after the Covid-19 pandemic that occurred in Indonesia because many workers were laid off due to the country's slowing economic growth. The government is trying to deal with the unemployment problem due to layoffs by establishing a Jaminan Kehilangan Pekerjaan (JKP) program managed by BPJS Ketenagakerjaan as an Unemployment Insurance product in Indonesia. The development and evaluation of this insurance product requires an actuarial assessment that includes demographic modeling as a review in describing regional employment risks to ensure fair treatment for all individuals. Demographic data in an area that includes the composition of the workforce can be the basis for demographic modeling, especially in the context of the Unemployment Insurance program. Almost every city or regency in Indonesia, including in Blitar City, is still experiencing unemployment problems so it is necessary to analyze the classification of the working status of residents in Blitar City which is able to provide an overview of employment conditions in the Blitar City area. The study conducted a work status classification analysis with a machine learning model on population data in Blitar City which was grouped into three categories, namely full employed, under employed, and unemployed. This analysis will use machine learning models with the Decision Tree Algorithm C4.5 method and Ordinal Logistic Regression to determine the best employment demographic classification model through a comparison of the accuracy level of the model formed. In predicting the category of work status, the model formed from the Decision Tree method of the C4.5 algorithm produces an accuracy value of 78.18%. On the other hand, the model formed from the Ordinal Logistic Regression method produced an accuracy value of 79.49%. Based on these results, the Ordinal Logistic Regression method is a method that produces a better model than the C4.5 Decision Tree Algorithm in predicting the classification of work status categories in Blitar City.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Algoritma C4.5, Klasifikasi, Regresi Logistik Ordinal, Status Kerja, Classification, C4.5 Algorithm, Employment Status, Ordinal Logistic Regression
Subjects:	Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Tianta Limara
Date Deposited:	31 Jan 2024 01:31
Last Modified:	31 Jan 2024 01:31
URI:	http://repository.its.ac.id/id/eprint/105778

Actions (login required)

View Item