Ecological vulnerability assessment serves as a prerequisite for ecological governance,yet evaluating large-scale ecological vulnerability remains challenging. To address thischallenge, this study integrates geological elements into ecological vulnerability assess-ment, taking Ruyuan Area in the Northern Guangdong Mountains, China, as a case study.The area faces ecological hazards such as land desertification and soil erosion, indicatingsevere governance challenges. This study selected 14 ecological vulnerability factors andconstructed assessment models based on Deep Neural Networks (DNNs) and Convolu-tional Neural Networks (CNNs). A total of 800 ecological vulnerability sampling pointswere obtained by combining field survey data with remote sensing imagery. The modelswere trained using binary vulnerability labels. The resulting continuous probability out-puts were then classified into five vulnerability levels using the natural breaks method togenerate the final ecological vulnerability map. It should be noted that the multi-level vul-nerability map represents graded probability-based differentiation rather than supervisedmulti-class prediction. Model performance was validated using three metrics: Area UnderReceiver Operating Characteristic Curve (AUC-ROC), Mean Absolute Error (MAE), andRoot Mean Square Error (RMSE). The CNN (AUC = 0.916) model outperformed the DNNmodel (AUC = 0.895). According to the CNN-based classification results, non-vulnerable,slightly vulnerable, mildly vulnerable, moderately vulnerable, and highly vulnerable areasaccounted for 36.19%, 22.85%, 14.24%, 12.31%, and 14.41% of the total area, respectively.High ecological vulnerability zones were concentrated in Daqiao, Luoyang, Dabu, andparts of Rucheng towns, with soil parent material and vegetation coverage identified asthe main contributing factors, among which parent material was the most important. Thisfinding underscores the notable impact of geological factors on local ecological vulnerability.Based on these results, nine ecological-geological subareas were delineated, and targetedecological protection and restoration recommendations were proposed. This study, em-ploying machine learning techniques, constructed an ecological vulnerability assessmentmodel incorporating geological elements, thereby providing scientific support for targetedecological governance in the study area.