AbstractIn recent decades developing nations have witnessed rapid urbanization with a massive inflow of rural migrants into their cities, creating large areas of informal settlements. However, city administrators have begun the regeneration of informal settlements resulting in gentrification and the involuntary resettlement of the informal settlement residents. Although a great deal of research has been conducted on regeneration projects, little is known about the resettlement outcomes of those removed. Therefore, it is essential to explore the spatial effect of urban regeneration and the relocation pathways of the forcibly removed tenants. This study examined the heterogeneity of residential relocation brought about by urban regeneration. With a support vector machine model applied under the proposed classification framework, six relocation patterns were identified in Baishizhou urban regeneration area, in Shenzhen, China. By analysing the variations in changes in housing security, neighbourhood and accessibility, the heterogeneity of relocation pathways revealed the relocation preference of different groups and the vulnerability of the residents. While the urban landscape's aesthetic quality may benefit from regeneration, the quality of life of many of those removed was shown to decrease. The results call for more social inclusion and sustainable urban regeneration, with efficient and targeted policy needing to be designed to protect the housing wellbeing of this vulnerable group.