Applied Spatial Analysis and Policy Geography, Planning and Development

Urban Housing Speculation Based on Housing Indoor Image Recognition: A Case Study of Kunming, China

Guo, Ziheng, Wang, Yang, Chu, Jun, Zheng, Haolei
Springer Nature · 2026-09-12 · DOI 10.1007/s12061-026-09998-7

Abstract

Housing speculation is an important factor affecting the stability and spatial differentiation of urban real estate markets, yet existing measurement approaches have difficulty capturing speculative differences at the intra-urban micro scale. Based on housing indoor image recognition, this study proposes a method that uses the actual occupancy rate of listed second-hand housing (during the listing period) as an inverse indicator of housing speculation and further constructs a Housing Speculation Index (HSI). Taking Kunming, China as a case study, this method is used to measure housing speculation at the block level and reveal its spatial pattern. The results show that the overall level of housing speculation in Kunming is moderate and exhibits a clear zonal gradient characterized by low speculation in the core area, moderate speculation in the main urban area, and high speculation in the suburban areas, with the suburbs becoming the main concentration of highly speculative housing. The findings demonstrate that the measurement framework based on housing indoor image recognition can effectively overcome the limitations of traditional macro-level indicators at the micro spatial scale, and provide a new perspective for identifying the intra-urban spatial pattern of housing speculation and formulating differentiated regulatory policies.

Read at publisher →