House Base-Price Prediction with Machine Learning Methods

Various factors like location, size, age, and condition must be considered to predict house prices accurately. This study utilizes historical sales data and demographic information in conjunction with machine learning algorithms, such as Random Forest, XGBoost, Support Vector Regression, and Artificial Neural Networks, to forecast home prices. The Kaggle dataset "House Sales in King County, USA" was used to test the proposed models. This research aims to help buyers find affordable properties and make informed decisions. Previous studies have shown that machine learning algorithms are more effective than traditional regression models in predicting house values. By utilizing artificial intelligence and machine learning, this study aims to improve the performance and effectiveness of predicting home prices. Our findings indicate that the ANN model has the highest performance results compared to the other models, with an performance rate of 0.785 before tuning and 0.887 after tuning. XGBoost comes in second with an performance rate of 0.886 after tuning, followed by Random Forest with an performance rate 0.855 after tuning. Finally, SVR is in fourth place with an performance rate 0.849 after tuning. Based on these results, we recommend using the ANN approach for estimating property prices in future research.