Integration of gustafson-kessel algorithm and ohonen's self-organizing maps for unsupervised clustering of seismic attributes

The goal of different methods for clustering of seismic attributes has been to analyze the discrimination ability of the chosen set of attributes. Kohonen clustering networks or Kohonen's self organizing maps are well known for cluster analysis (unsupervised learning). This class of algorithms is a set of heuristic procedures that suffers from some major problems. In this paper we propose the use of an unsupervised method by integrating the fuzzy c-means clustering and Gustafson-Kessel algorithms into the learning rate and updating strategies of the Kohonen clustering network. Using a new fuzzy calibration method, the different attribute classes are calibrated to lithology classes and the appropriate attributes and classes are determined. In this paper we propose a robust clustering algorithm which can be quality-controlled by using fuzzy modeling. That means using the clustering results, the available but limited log data is rebuilt after clustering to see the performance of the clustering technique. Classification and modeling results tested on a real data set show reasonable accuracy when compared to well logs.