Abstract: As one of the main aquifers of the Jurassic coal seam in northern Shaanxi, the weathered bedrock aquifer seriously threatens the safe production of mines. Therefore, it is crucial to scientifically and accurately predict the water richness of weathered bedrock aquifers. Taking the south wing of Ningtiaota Coal Mine as an example, according to the borehole data, five factors were selected as evaluation indexes, including the thickness of the weathered bedrock, the lithology combination index, the weathering index, the core taking rate and the buried depth, and a support vector machine (SSA−SVM) model based on Sparrow algorithm optimization was constructed to predict the water richness of the weathered bedrock aquifer. The results show that the SSA−SVM prediction model shows higher accuracy than the unoptimized support vector machine and the support vector optimized by particle swarm optimization. The weathered bedrock in the study area is generally moderately water−rich and spatially unevenly distributed. The water evacuation data of the S1231 working face and the data of the previous water inrush points are verified, which is consistent with the partition results of the SSA−SVM prediction model. The SSA−SVM prediction model is feasible to predict the water richness of weathered bedrock aquifers, which has reference significance for predicting the water richness of weathered bedrock aquifers in Ningtiaota Coal Mine and surrounding mines.