Full text loading...
-
Integration and zonation of geophysical data sets with skewed histographic data distribution using Gustafson-Kessel cluster analysis
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, EGM 2010 International Workshop, Apr 2010, cp-165-00017
Abstract
We employ the fuzzy Gustafson-Kessel (GK) cluster algorithm for rapid and objective integration of a disparate geophysical data base comprising airborne radiometric and magnetic as well as ground-based gravity data. The data base has previously been integrated using the fuzzy c-means (FCM) cluster algorithm, which required a significant amount of data preprocessing to meet the particular requirements of the FCM algorithm. Since the GK algorithm is more robust with regard to data scaling and histographic data distribution, we repeatedly integrate the available data base while lowering the amount of preparatory data processing and scaling. The final results closely match those previously obtained by the FCM-based integration which had already been associated to known geological information in the survey area.