The Implementation of Probabilistic Neural Network Algorithm for Classification of Family Hope Program in Pekanbaru City

Abstract

Abstract The poverty rate in Indonesia continues to increase, especially in Pekanbaru city. It was recorded in 2017 the poverty in Pekanbaru City reached 176.908 inhabitants. The poverty can be seen from the Education Factor, Economic Factor, Health Factor and Infrastructure Factor. This aim of this study is to classify beneficiary of poverty from education and health factors using the PNN Algorithm. The criteria used for the classification of education classes and health classes include elementary, junior high, high school, toddlers and pregnant women. The data used in this study were from the Harapan family program in Pekanbaru City. In clustering training data and test data, K-Means Algorithm was used. The results of the clustering are 3,543 test data and 1,520 testing data with DBI value of 0.194 and the result of calculation of Probabilistic Neural Network algorithm with accuracy value is 99.07%. In testing the algorithm using the confucion matrix method, the recall value is 99.30% and the precision value is 97.81%. The high level of health and education is an important factor in the development and progress of an area.