Abstract: To overcome the limitations of traditional resource evaluation methods that solely rely on either expert-driven or data-driven models, and to improve the objectivity and result reliability of the evaluation, this paper conducts a study taking the potential of deep geothermal resource in southern Jiangsu Province as an example. Following a systematic analysis of geological settings and distribution characteristics, seven key factors reflecting the formation mechanism and characteristic markers, which covering the core elements of “source-connectivity-reservoir-cap” of deep geothermal resources, were selected to establish an evaluation index system, including crustal thickness, Curie depth, distance from deep large faults, distribution of medium-acidic magmatic rocks, terrestrial heat flow, geothermal gradient, and caprock thickness. Specifically, the weights of expert-driven and data-driven indicators were calculated using the G1 method and entropy weight method, respectively; subsequently, game theory was employed to determine the combined weights of the indicators, and a data-expert dual-driven model was established to quantitatively evaluate the distribution potential of deep geothermal resources in southern Jiangsu Province. The results indicate that the high-potential areas of deep geothermal resources in the study area are concentrated in the southern part of Nanjing and the northeastern part of Suzhou, distributed in the NE and NW directions, which are controlled by regional fault structures and their intersection zones. From the perspective of quantitative evaluation and calculation, these results are consistent with the existing research conclusions of deep geothermal fields, and give an explanation for the traditional geological understanding that deep geothermal resources in southern Jiangsu are mainly structural fracture type. At same time, it also indicates that the dual-driven model effectively integrates expert experience and the objective characteristics of data, providing an effective method for the potential evaluation of deep geothermal resources.