Self-organizing maps for geoscientific data analysis: geological interpretation of multidimensional geophysical data

A self-organizing map (SOM) approach has been used to provide an integrated spatial analysis and classification of airborne geophysical data collected over the Brazilian Amazon. Magnetic and gamma ray spectrometric data were used to extract geophysical signatures related to the spatial distribution of rock types and to produce a geologic map over the prospective Anapu-Tuerê region. Particular emphasis was given to discriminating and identifying rock types, and the processes related to gold mineralization, which are known to occur in the Anapu-Tuerê region. SOM was able to identify and map distinctive geophysical signatures related to the various geologic units identified on the published geologic map. Furthermore, SOM was able to identify and enhance very subtle signatures derived jointly from the magnetic and gamma ray spectrometric data that could be related to geologic processes present in the area. These results demonstrate the effectiveness of using SOM as a tool for geophysical data analysis and for semiautomated mapping in regions such as the Amazon.

Download Full-text

Self-Organizing Maps In Airborne Geophysical Data Applied To Geological Mapping In Amazonian Region

10.3997/2214-4609-pdb.195.2064_evt_6year_2009 ◽

2009 ◽

Author(s):

Cleyton De Carvalho Carneiro ◽

Stephen James Fraser ◽

Alvaro Penteado Crósta ◽

Adalene Moreira Silva ◽

Carlos Eduardo De Mesquita Barros

Keyword(s):

Geological Mapping ◽

Geophysical Data ◽

Self Organizing Maps ◽

Airborne Geophysical Data ◽

Amazonian Region ◽

Self Organizing

Download Full-text

Exploratory Data Analysis Techniques Applied to Hydrological Time Series. I: Self-Organizing Maps

10.13031/2013.24291 ◽

2008 ◽

Author(s):

Diego A Rivera ◽

José L Arumí ◽

Mario F Lillo

Keyword(s):

Time Series ◽

Data Analysis ◽

Exploratory Data Analysis ◽

Self Organizing Maps ◽

Analysis Techniques ◽

Exploratory Data ◽

Self Organizing

Download Full-text

Self-Organizing Maps and Their Applications to Data Analysis

10.2172/1566795 ◽

2019 ◽

Author(s):

R Ponmalai ◽

C Kamath

Keyword(s):

Data Analysis ◽

Self Organizing Maps ◽

Self Organizing

Download Full-text

Generalized EmbedSOM on quadtree-structured self-organizing maps

F1000Research ◽

10.12688/f1000research.21642.2 ◽

2020 ◽

Vol 8 ◽

pp. 2120

Author(s):

Miroslav Kratochvíl ◽

Abhishek Koladiya ◽

Jiří Vondrášek

Keyword(s):

Data Analysis ◽

Dimensionality Reduction ◽

Single Cell ◽

Manifold Learning ◽

Generating Functions ◽

The Self ◽

Reduction Algorithm ◽

Self Organizing Maps ◽

Learning Techniques ◽

Self Organizing

EmbedSOM is a simple and fast dimensionality reduction algorithm, originally developed for its applications in single-cell cytometry data analysis. We present an updated version of EmbedSOM, viewed as an algorithm for landmark-directed embedding enrichment, and demonstrate that it works well even with manifold-learning techniques other than the self-organizing maps. Using this generalization, we introduce an inwards-growing variant of self-organizing maps that is designed to mitigate some earlier identified deficiencies of EmbedSOM output. Finally, we measure the performance of the generalized EmbedSOM, compare several variants of the algorithm that utilize different landmark-generating functions, and showcase the functionality on single-cell cytometry datasets from recent studies.

Download Full-text