Determination of the Best Accuracy Model for Predicting Average Years of Schooling using the Fletcher Reeves Algorithm

Abstract

The average length of schooling is an important and significant factor in looking at the quality of an individual human being, with increasing the quality of human resources it can increase access to decent work which
also promises a stable economic income, and to some extent affects the economy in a country. Therefore, a prediction was made. This prediction method uses the Fletcher Reeves algorithm which is an artificial neural network algorithm method for data prediction. However, this paper does not discuss the results of the prediction, but discusses the ability of the Fletcher Reeves neural network algorithm to predict data. The research dataset used in this study is data on the average length of schooling in North Sumatra Province from 2015-2020, this dataset was taken from BPS North Sumatra. The data is then formed into 5 models, namely 2-10-1, 2-15-1, 2-20-1, 2-25-1, 2-30-1. -30-1 with an MSE value of 0.000430727. With these results the 2-30-1 architectural model gets the lowest score, so it can be concluded that the model can be used to predict the average length of schooling in North Sumatra
Province.