A single-valued neutrosophic affinity propagation approach for engineering geological zoning of open-pit mine slopes

Abstract

Engineering geological zoning is an important foundation for assessing the stability of open-pit mine (OPM) slopes. Because of the complexity and uncertainty of geological conditions, this work brings great challenges to mining engineers and researchers. This paper presents a single-valued neutrosophic affinity propagation (SVNS-AP) approach for engineering geological zoning of OPM slopes. The approach utilizes the concept of a single-valued neutrosophic set (SVNS) to express the inconsistent and indeterminate information present in the influencing factors of engineering geological conditions through truth, indeterminate and falsity membership functions. Then, the similarity measure of SVNS is integrated into the affinity propagation (AP) algorithm to calculate the degree of similarity between the data points. Finally, the modified silhouette index is used to evaluate the clustering results and decide the optimal number of clusters. The practical application results of the engineering geological zoning of Lanping lead-zinc OPM demonstrate that the SVNS-AP method is an effective way for engineering geological zoning of OPM slopes in the uncertain environment. Clustering results based on datasets in the literature and the UC Irvine Machine Learning Repository show that the proposed method can be used as a general clustering algorithm.