Prediction of Social Assistance Recipients Using K-Nearest Neighbor (KNN) (Case Study: West Balikpapan District)
Muliady Faisal (a*), Rahayu Eka Pratiwi (b), Itva Aliva (b)

Department of Mathematics, Institut Teknologi Kalimantan

Jl. Soekarno Hatta No.KM 15, Karang Joang, Kec. North Balikpapan, Balikpapan City, East Kalimantan 76127

Abstract

Poverty is the inability from the economic side to fulfill basic needs and non-food which is measured from the expenditure side. Service Social has implemented a social assistance system for underprivileged citizens, the poor, or unable to reduce poverty levels. There are some things to consider such as the economic condition of the residents or having several
categories namely, work, income, and dependents. Based on this problem, researchers want to predict social assistance recipients using K-Nearest Neighbor (KNN), especially Balikpapan District West. From the results of research that has been conducted by researchers, it can be concluded that the level of accuracy of using K-Nearest Neighbor has good accuracy. Accuracy results are carried out by 3 tests namely obtained an accuracy of 74.2% with a comparison of training and testing data 70%: 30% with a value of k = 9, a comparison of training and testing data 80%: 20% obtained accuracy of 82% with a value of k = 10, and a comparison of training data and testing 90%: 10% obtained an accuracy of 83% with a value of k = 10.

Keywords: Poverty- Social Assistance- K-Nearest Neighbor

Topic: COMPUTATIONAL SCIENCES