Self-organizing map algorithm for assessing spatial and temporal patterns of pollutants in environmental...

Area of Science:

• Environmental Science
• Data Science
• Cheminformatics

Background:

• Assessing anthropogenic environmental impact requires understanding pollutant distribution.
• Chemometric methods are vital for environmental health data exploration.
• Unsupervised learning, particularly Self-Organizing Maps (SOMs), offers powerful tools for complex environmental data analysis.

Purpose of the Study:

• To review the Self-Organizing Map (SOM) algorithm for environmental data analysis.
• To detail SOM's application in identifying spatial and temporal pollution patterns.
• To provide guidance on SOM implementation, interpretation, and reporting for reproducibility.

Main Methods:

• Description of the Self-Organizing Map (SOM) algorithm and its key parameters.
• Explanation of SOM output features for data mining and pattern recognition.
• Integration of SOM with clustering algorithms to enhance interpretation.

Main Results:

• SOMs effectively handle non-linear problems in environmental data.
• Combined SOM and clustering reveal intricate spatial and temporal pollution patterns.
• The review identifies software tools and provides visualization strategies for SOM results.

Conclusions:

• Self-Organizing Maps offer a robust approach for environmental pollution pattern analysis.
• Merging SOM with clustering significantly improves the interpretability of environmental data.
• Standardized reporting of SOM methods is crucial for research comparability and reproducibility.