Survey of AI Methods for the Purpose of Geotechnical Profile Creation

The goal of this paper is to present methodology of unsupervised learning application in geotechnical data categorization. Geotechnical layers identification is conducted based on measurement from the Dilatometer of Marchetti Test taken at the campus of Warsaw University of Life Sciences. To cluster data, the Ant Clustering Algorithm, k-means, fuzzy sets and Self Organizing Map algorithms were introduced. All methods are adjusted to the presented problem and their efficiency compared. The paper is concluded with comments about the applications of computer intelligence methods for the geotechnical data analysis.