ANALISIS PREDIKSI HARGA RUMAH DI JABODETABEK MENGGUNAKAN MULTIPLE LINEAR REGRESSION

Abstract

The economic and infrastructure development in the Jabodetabek area has increased the rate of population growth in the region continues to increase. This attracted the attention of property agents in the housing price competition in the area, considering that the demand for housing will also continue to increase. There are several kinds of specificities and house prices in the Jabodetabek area, so an appropriate analysis is needed to get the desired house in this study the researcher implements the Multiple Linear method Regression to predict house prices with the desired specifications. This study used a total dataset of 3553 data with 11 variables obtained from the Kaggle.com. website The data will be through the pre-pronce process of data before training the model. Furthermore, at the evaluation stage using a test matrix, namely CC (Coefficient of Correlation), MAE (Met Absolute Error), RMSE (Root Met Square Error) is used to assess the performance of the model. Based on the results of the analysis from the dataset were 965 data, where the data was divided into two parts, namely, 80% as training data and 20% as testing data from the analysis results were obtained at an accuracy rate of 0.85 or 85% with an error rate of MAE of 375428909.51 and the RMSE level of 516385,50.