PEMODELAN REGRESI LINEAR PADA PREDIKSI JUMLAH PENDUDUK DI KOTA TANJUNGPINANG MENGGUNAKAN ALGORITMA GENETIKA

Abstract

Tanjungpinang city is one of the growing cities in Indonesia which growth show no sign of stopping with population increase every year. Population is one of the important indicators in city management to prevent unemployment and crime increase. Thus, it’s necessary to develop a program that can estimate the population growth in Tanjungpinang city. In this research, this problem will be modeled with linear regression that consists of the independent variable (X) and the dependent variable (Y). Coeffiecient regression would be generated using the concept of genetic algorithms. In the reproduction process using the extended intermediate crossover and random mutation, for the selection process using an elitism selection models. With final result in prediction using Linear Regression Modelling with sum of Error 1790005.143, best popSize values is 200, the best combination of crossover rate with mutation rate is 0.6:0.4, and the optimized number of generation is 100.