Accurately assessing the risk level of landslides is of great significance for disaster prevention and mitigation. In this research, a method of landslide stability targeting analysis considering displacement traction is proposed, which considers the traction correlation characteristics of displacement vectors between monitoring points and carries out the optimal grid unit division of landslides. Through the Improving Sparrow Search Algorithm (ISSA) combined with the K-means clustering algorithm, clustering analysis was executed based on the displacement characteristics of the monitoring points and combined with the optimal grid unit to form the target area, and the assessment sample set of the corresponding risk category of the landslide unit was constructed. Further, on the basis of the landslide target area, the stability targeting analysis of 11 influencing factors, such as surface roughness, elevation, slope, and so on, was carried out by using the Stacking model. The experimental results show that compa