3D pseudo-lithologic modeling via iterative weighted k-means++ algorithm from Tengger Desert cover area, China

The bedrock beneath the Tengger Desert is covered by Quaternary deposits, making it difficult to directly observe underlying geological information using traditional methods. In areas with limited prior information, employing geophysical methods obtain deep-seated constructing a multi-source dataset, and performing three-dimensional modeling can significantly enhance our understanding of underground structures. Cluster analysis fundamental unsupervised machine learning technique employed in data mining investigate structure within feature space. This paper proposes an iterative weighted distance-based extension k-means clustering algorithm, referred as Iterative Weighted Distance K-means (IW k-means++) algorithm. It incorporates farthest distance method select initial centroid, performs centroid updates based on distance, dynamically adjusts weights during training. Davies-Bouldin index shows that performance IW ++ algorithm better than K-Meme 3D pseudo-lithology modeling.