Penerapan Metode Jaringan Syaraf Tiruan Backpropagation untuk Prediksi Tingkat Kemiskinan di Provinsi Jawa Tengah

Dian, Finaliamartha (2021) Penerapan Metode Jaringan Syaraf Tiruan Backpropagation untuk Prediksi Tingkat Kemiskinan di Provinsi Jawa Tengah. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text Cover.pdf Download (546kB)
	Text Abstract.pdf Download (86kB)
	Text Abstrak.pdf Download (99kB)
	Text BAB I.pdf Download (157kB)
	Text BAB II.pdf Download (723kB)
	Text BAB III.pdf Download (166kB)
	Text BAB IV.pdf Restricted to Registered users only Download (664kB) | Request a copy
	Text BAB V.pdf Download (101kB)
	Text Daftar Pustaka.pdf Download (100kB)
	Text Lampiran.pdf Restricted to Registered users only Download (586kB) | Request a copy

Abstract

Poverty is a problem that should be seen as a complex (multidimensional) social problem. One of the objectives of the development plan is to improve people's welfare by reducing poverty and unemployment. Based on data from the Central Statistics Agency, the percentage of national poverty in March 2019 was 9.41 percent. Meanwhile, Central Java Province has a poverty rate higher than the national poverty rate, which is 10.8 percent. High levels of poverty can lead to crime, high unemployment, social, political, and other chaos. Therefore, this study aims to analyze the level of poverty by determining the appropriate model which can then be used to predict poverty levels according to districts/cities in Central Java Province. The data used comes from the Central Java Provincial Statistics Agency based on districts/cities in the form of percentage data from 2010 to 2019 which consists of input variables, namely data on economic growth rates, open unemployment rates, human development indexes, and output variables, namely poverty levels according to district/city. The method used in this study is the Backpropagation Neural Network. The Backpropagation Artificial Neural Network has a good performance in solving problems, one of which is prediction problems. Based on the best architectural model produced in this study, the 3- 35-1 architectural model can produce an accuracy rate of 95.2% using MSE in the testing process using test data. So it can be concluded that the Backpropagation Neural Network by applying the appropriate network architecture model can produce a good level of accuracy which can then be used as an alternative to predict poverty levels according to regencies/cities in Central Java Province in the future. Kata kunci : Artificial Neural Network, Backpropagation, Poverty Level, Prediction, Matlab

Actions (login required)

View Item