Estimating the price of residential properties based on the optimal support vector machine

In developed economies, taxes based on residential property prices make a significant contribution to the sustainable income of the city managers. Therefore, estimating the price of residential properties is very important for economic purposes. Estimating the price of residential properties is a complex nonlinear, and multivariate problem. In this study, a hybrid method of support vector machine (SVM), genetic algorithm (GA) and particle swarm optimization (PSO) was used to estimate the price of residential properties. The support vector machine has been proven to be a powerful and robust algorithm for regression and classification. However, selecting the most appropriate hyper-parameters of this algorithm is a significant problem for its implementation. For hybrid SVR algorithms with PSO and GA, the mean absolute error is respectively ۱۰.۱۳% and ۱۰.۱۴%, based on the results of this study.

Keywords:

Residential Property Price Estimation , Support Vector Machine , Genetic Algorithm , Particle Swarm Optimization