House price prediction using machine learning: A case study in Seattle, U.S.

Abstract

In recent years, predicting housing prices has become a prominent research topic. The motivation behind this research is the lack of precise analysis and comprehensive comparison of different machine learning models for Seattle housing prices. To address this issue, this paper utilizes a dataset obtained from Kaggle, consisting of Seattle housing prices. This study focuses on analyzing and comparing the performance of different machine learning models in predicting housing prices in Seattle. After preprocessing the data, eight different machine learning models are applied to predict housing prices in Seattle. A comprehensive comparison of these models is conducted to analyze their differences. The experimental results show that the differences in performance among the models are not substantial. However, StackingAveragedModels emerges as the top-performing model, with an RMSLE of 0.2328 and an R2 of 0.7771. These findings contribute to a better understanding of the predictive capabilities of different machine learning models for Seattle housing prices.