A Prediction Study on Archaeological Sites Based on Geographical Variables and Logistic Regression—A Case Study of the Neolithic Era and the Bronze Age of Xiangyang

Archaeological site predictive modeling is widely adopted in archaeological research and cultural resource management. It conducive to excavation reveals the progress of human social civilization. Xiangyang City focus this paper. We selected eight geographical variables as influencing variables, which are elevation, slope, aspect, micro-landform, slope position, plan curvature, profile distance from water. With them, we randomly obtained 260 non-site points at ratio 1:1 between based on excavated sites constructed a sample set geospatial data logistic regression (LR). Using 10-fold cross-validation, trained tested model select best samples. Thus, quantitative relationship was established. As result, Area Under Curve (AUC) LR 0.797 its accuracy 0.897 study. A detector unveils that three Distance water, elevation Plan Curvature top chart. The under highly stable accurate. constitute crucial archaeological.