Klasterisasi Kasus Kemiskinan di Indonesia pada Hasil Peramalan Backpropagation Neural Network

Pradnyandari, Ni Putu Putri Marinda (2023) Klasterisasi Kasus Kemiskinan di Indonesia pada Hasil Peramalan Backpropagation Neural Network. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06311940000049-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (3MB) | Request a copy

Abstract

Kemiskinan merupakan permasalahan sosial yang cukup kompleks dan sulit untuk hilang. Banyak negara sudah memfokuskan tujuan kenegaraan mereka untuk menanggulangi kemiskinan. Hal ini dibuktikan dengan lahirnya Sustainable Development Goals (SDGs) yang disusun oleh Perserikatan Bangsa-Bangsa (PBB) dan mengharuskan Indonesia turut andil dalam penanggulangan kemiskinan. Namun, jika kita lihat berdasarkan data, menurut data Badan Pusat Statistik, angka kemiskinan di Indonesia masih mengalami peningkatan di bulan September 2022 dibandingkan dengan periode sebelumnya, Maret 2022. Berdasarkan permasalahan tersebut, solusi yang dapat ditawarkan untuk membantu pemerintah dan masyarakat dalam menanggulangi kemiskinan adalah dengan melakukan peramalan sebagai landasan dalam mengambil kebijakan. Peramalan ini menggunakan metode Backpropagation Neural Network dan hasil data peramalan dilakukan clustering menggunakan metode K-Means. Objek penelitian yang digunakan adalah data kemiskinan 34 provinsi di Indonesia dari tahun 2015 sampai 2022 yang terdiri dari variabel tingkat kemiskinan, PDRB harga konstan, tingkat pengangguran terbuka, dan rasio gini. Data berupa data semester dengan total jumlah data di setiap variabelnya sebanyak 544 data. Dalam pemodelan BPNN, didapatkan model terbaik dengan jaringan BPNN (4-6-1) dengan hasil MAPE dan MSE sebesar 7,14% dan 0,00000492. Hasil peramalan berdasarkan model BPNN terbaik kemudian dilakukan klasterisasi. Clustering K-Means menghasilkan 3 klaster dengan karakteristik yang berbeda. Jumlah provinsi yang masuk ke dalam klaster 1 sebanyak 8 provinsi, klaster 2 sebanyak 5 provinsi, dan klaster 3 sebanyak 21 provinsi.
================================================================================================================================
Poverty is a complex social problem that is difficult to eliminate. Many countries have focused their state goals on tackling poverty. This is evidenced by the birth of the Sustainable Development Goals (SDGs) compiled by the United Nations (UN) and requires Indonesia to take part in poverty reduction. However, if we look at the data, according to the Central Bureau of Statistics, the poverty rate in Indonesia still increased in September 2022 compared to the previous period, March 2022. Based on these problems, a solution that can be offered to help the government and society in reducing poverty is to conduct forecasting as a basis for making policies. This forecasting uses the Backpropagation Neural Network method and the results of the forecasting data are clustering using the K-Means method. The research object used is poverty data for 34 provinces in Indonesia from 2015 to 2022 which consists of variables of poverty rate, GRDP at constant prices, open unemployment rate, and Gini ratio. The data is in the form of semester data with a total of 544 data in each variable. In BPNN modeling, the best model is obtained with the BPNN network (4-6-1) with MAPE and MSE results of 7.14% and 0.00000492. Forecasting results based on the best BPNN model are then clusterized. K-Means clustering produces 3 clusters with different characteristics. The number of provinces included in cluster 1 is eight provinces, cluster 2 is five provinces, and cluster 3 is 21 provinces.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Forecasting, Poverty, Backpropagation Neural Network, K-Means Clustering, Peramalan, Kemiskinan, Clustering K-Means
Subjects:	Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics) T Technology > T Technology (General) > T174 Technological forecasting T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Ni Putu Putri Marinda Pradnyandari
Date Deposited:	26 Jul 2023 02:49
Last Modified:	26 Jul 2023 02:49
URI:	http://repository.its.ac.id/id/eprint/99719

Actions (login required)

View Item