<p>Land subsidence not only poses a serious threat to urban construction and the safety of people’s lives and property, but also brings serious damage to urban infrastructure. Therefore, solving the problem of land subsidence is essential to ensure the sustainable development of cities and the quality of life of people. At present, the prediction of the future trend of land subsidence is often limited by a single prediction model or technical means, and its prediction results have certain uncertainties and limitations. Therefore, this paper takes Liaocheng City, Shandong Province as the study area, and uses a variety of methods to study the development status and future development trend of land subsidence in the study area in detail based on the relevant historical and current data of land subsidence and the actual monitoring data. The results show that if the current mining volume is maintained, the maximum regional subsidence will be 1.3 times that of the current ten years. The water level of shallow groundwater has rebounded significantly under the condition of 30% reduction of extraction, and the recovery rate of deep groundwater has also increased to varying degrees, indicating that reducing the exploitation of groundwater is one of the effective measures for the prevention and control of land subsidence. The machine learning model is used to predict the development trend of land subsidence in the study area, and the land subsidence in the study area will show a trend of further development, and the subsidence will develop more rapidly near the subsidence funnel. This study can provide scientific guidance for the prevention and control of regional land subsidence.</p>