Backpropagation Neural Network Method For The Classification of Districts/Cities Based On Macro Socio-Economic Indicators In The Province Of South Sulawesi

Rais, Zulkifli
Sudarmin
Syahputra, Akbar

Quantitative Economics and Management Studies
Vol. 6 No. 2 (2025)

Classification is a way of grouping objects based on the characteristics possessed by the objects of classified. One of the developing classification methods is the backpropagation neural network. This study aims to look at the descriptive and classification results of the District/City Macro Socioeconomic Indicators in South Sulawesi Province. The data set comprises 24 observations with 9 variables, namely population density, poverty line, Gini ratio, open unemployment rate, life expectancy, average length of schooling, labor force participation rate, life growth rate, and GRDP at current prices. A model with a total of 9 hidden layers and a learning rate of 0.002 is obtained with an accuracy of 70%, precision of 70%, recall of 100%, and F1 score of 87%.