关键词: 机器学习, K 邻近域算法, 全域地质特征预测, 地质建模, 钻孔数据

Abstract: To solve the problem that traditional modeling methods can’t deeply excavate the original drilling data
or make sufficient use of geological data, this paper proposes an improved KNN (K-Nearest Neighbor) algorithm.
It is a spatial adaptive interpolation fitting algorithm that is developed based on the original KNN algorithm. It
incorporates automatic selection of k-values based on different geological layers and utilizes the original geological
data for further analysis. Taking a railroad investigation project data as a data source, importing this data into the
improved KNN algorithm model and running it, we successfully obtained the characteristic k values of each stratum
in the site and accomplished the geological modeling. By comparing the actual borehole and the borehole predicted
网络首发时间：2023-10-19 18:21:55
网络首发地址：https://link.cnki.net/urlid/44.1745.U.20231019.1532.002
by original KNN and improved KNN, we found that the improved KNN algorithm is more accurate in predicting
the thin layers, and the overall accuracy is higher. Compared with the original KNN algorithm and other common
classification algorithms, the improved algorithm can obtain better stratum prediction results, especially for the thin
stratum, and obtain higher precision and accuracy, which can better guide the prediction of underground 3D space.

Key words: Machine Learning, K-Nearest Neighbor, Prediction of Global Geological Features, Geological
modeling, Drilling data