Abstract

To handle the increasing volume of seismic data, machine learning methods, specially unsupervised classification techniques, have emerged as a valuable tool. This research paper focuses on the application of unsupervised classification methods, specifically Principal Component Analysis (PCA) and Kohonen Self-organizing maps, for seismic facies analysis. Through case studies from the F3 Block of the Netherlands, the effectiveness and potential of these techniques are explored. Mathematically derived Seismic attributes were used as an input for clustering and recognizing seismic facies. By capturing the variability within seismic data, these methods reveal crucial insights into subsurface geological features. This paper highlights the importance of efficient data reduction algorithms for optimizing the analysis of seismic facies.