A SOM-view of oilfield data: A novel vector field visualization for Self-Organizing Maps and its applications in the petroleum industry

Self-Organizing Maps are a prominent tool for exploratory analysis and visualization of high-dimensional data. We propose a novel method for visualizing the cluster structure and coherent regions of the Self-Organizing Map that can be displayed as a vector field on top of the map lattice. Concepts of neighborhood and proximity on the map is exploited to obtain a representation where arrows point to the most similar region. The method is especially useful for large maps with a high number of map nodes. In our experiments, we visualize a data set that stems from applications in the petroleum industry, and show how to use our method to maximize the gas output. Key Words: Self-Organizing Maps, Knowledge Visualization, Petroleum Industry Category: I.5.3