Abstract: Objective Human soil erosion is an important form of soil and water loss. Zhaoqing City is located in the Guangdong Hong Kong Macao Greater Bay Area, with frequent human activities. Scientific assessment of the risk level of human soil erosion of key objects in the region can provide scientific basis for water administrative departments to implement precise supervision. Methods Taking the human induced soil erosion disturbance map of Zhaoqing City in 2024 as the research object, a sample dataset was constructed based on key indicators such as disturbance plot area, perimeter, shape index, slope ratio, average slope, height difference, elevation mean, and distance from rivers and roads. Random forest （RF）, decision tree （DT）, logistic regression （LR）, and support vector machine （SVM） models were trained and compared.Results The random forest model performed the best on the test set, with an accuracy of 89.4%, a macro F1 score of 0.883, and a Kappa coefficient of 0.834, demonstrating good classification performance and generalization ability. Conclusion The constructed multi parameter indicator system