Abstract
Housing is among the most pressing issues in China. Researchers are eager to identify housing property's internal and geographic factors influencing residential property prices. However, few studies have examined the relationship between social media users' point of interest (POI) data and house prices using big data. This paper presents a machine learning model for regression analysis to reveal the relationship between housing prices and check-in POI density in Futian District, Shenzhen. The results show that our proposed price prediction model using additional features based on POI data proved to provide higher prediction accuracy. Our results indicate that incorporating POI features based on current feeds from location-based social networks can provide more up-to-date estimates of housing market price trends.