An enriched K-means clustering method for grouping fractures with meliorated initial centers

© 2014, Saudi Society for Geosciences. An enriched K-means clustering method for grouping fractures with meliorated initial cluster centers is proposed. Selection of the initial cluster centers is based on a gathering degree function and a hierarchical clustering method. A simplified Xie-Beni cluster validity index is applied to determine the optimal number of clusters automatically. The effectiveness of the proposed clustering method is demonstrated by using synthetic data and field data compiled from literatures. The gathering degree concept used in the density-based method is helpful in seeking suitable initial cluster centers. It alleviates the influence from the selected arbitrary threshold to get a more stable clustering result. The procedure to automatically remove fractures belonging to the obtained cluster center with the hierarchical clustering method is more natural than using the pre-defined radius. Additional advantage of the algorithm is that its convergence is fast, and it can be easily implemented for geological mapping result analysis.