PREDICTION OF THE POOR RATE K-MEANS AND GENERALIZED REGRESSION NEURAL NETWORK ALGORITHMS (CASE STUDY: NORTH SUMATRA PROVINCE)

Abstract

Poverty reduction is a crucial issue and the primary The North Sumatra Provincial government's main concern is lowering the poverty rate, which is a crucial issue. The Province of North Sumatra in Indonesia, one of many nations affected by the Covid-19 pandemic, is particularly troubled economically. In this study, poverty levels were mapped using the K-Means algorithm, and GRNN was then utilized for modeling and prediction. The data source used is time series data from 2010 to 2020 from the Central Statistics Agency (BPS), which includes variables X covering population, health, education, unemployment, and asset ownership and variable Y representing poverty level. The goal of this study is to choose the best model for estimating poverty levels in North Sumatra Province. The districts and cities of Deli Serdang and Medan have the greatest rates of poverty, according to the K-means algorithm's mapping of poverty levels. Additionally, the results of the predicting produced MSE values of 0.004659 and RMSE values of 0.00002108. The value of the smoothness parameter is 0.01.