论文信息 - Self-Organizing Maps: Learning for Exploratory and Predictive Modelling

Self-Organizing Maps: Learning for Exploratory and Predictive Modelling

Self-organizing maps (SOMs) involve using machine learning methods to visualize different patterns in data and to determine the relationship between experimental measurements and samples. SOMs were first reported by the Finnish professor Teuvo Kohonen, and is sometimes called a Kohonen map [1,2]. For 30 years the SOMs have been widely employed for visualization of relationships between samples. Unsupervised SOMs as traditionally employed are used primarily for exploratory data analysis to reveal relationships between samples in data. They allow visualization of a large number of samples in limited space. However, it is also possible to employ SOMs for classification purposes whereby an additional vector of class information is included in the training.

R. G. Brereton | R. Brereton

[1] Teuvo Kohonen,et al. Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.

[2] R. Brereton,et al. Self-organizing map quality control index. , 2010, Analytical Chemistry.

[3] R. Brereton,et al. Supervised self organizing maps for classification and determination of potentially discriminatory variables: illustrated by application to nuclear magnetic resonance metabolomic profiling. , 2010, Analytical chemistry.