A Geographic Feature Integrated Multivariate Linear Regression Method for House Price Prediction
- DOI
- 10.2991/assehr.k.201214.522How to use a DOI?
- Keywords
- House price prediction, Multiple linear regression models, Geographic feature integrated multivariate linear regression method, Real-world case
- Abstract
Housing price prediction is of great significance in financial real estate investment and urban construction planning. Multiple linear regression models are commonly used for housing price prediction. However, traditional methods are mostly focused on the characteristics of the houses themselves, without or little considering the features of the surroundings. The features of the surroundings are also important for house price prediction. Motivated by these, we propose a geographic feature integrated multivariate linear regression method for house price prediction. Especially, the Zip Code is chosen as the additional geographic feature for its convenience to obtain. Then the integrated features are used to learn the multivariate linear regression model. We conduct an extensive experiment on the real-world case of the King County area and compare our method linear regressions. The results verified the effectiveness and superiority of our model.
- Copyright
- © 2020, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Yuhang Mao AU - Ruili Yao PY - 2020 DA - 2020/12/16 TI - A Geographic Feature Integrated Multivariate Linear Regression Method for House Price Prediction BT - Proceedings of the 2020 3rd International Conference on Humanities Education and Social Sciences (ICHESS 2020) PB - Atlantis Press SP - 347 EP - 351 SN - 2352-5398 UR - https://doi.org/10.2991/assehr.k.201214.522 DO - 10.2991/assehr.k.201214.522 ID - Mao2020 ER -